Littérature scientifique sur le sujet « Visione Artificiale ed Image Processing »

The object of mapping is to produce a "correct" relational model of the terrain. Its assumptions are that the objects in the world to be mapped are real and objective, and that they enjoy an existence independent of the cartographer; that their reality can be expressed in mathematical terms; that systematic observation and measurement offer the only route to cartographic truth; and that this truth can be independently verified. -- J. B. Harley, "Deconstructing the Map" Cartography, in its pragmatic operation under these assumptions, avoids almost all of the problems of representation with which cultural studies is only too familiar. Maps are representations, and all representations, even "scientific" ones, are cultural signs rather than truths produced in an ideological vacuum. The notion that mathematics, as a tool of Renaissance rationality, allowed real, objective detachment from an object of study was slowly absorbed into the cartographic tradition of Europe about four hundred years ago. "From at least the seventeenth century onward there was an epistemic break in activities such as cartography and architecture, and European map makers increasingly promoted what we would describe today as a standard scientific model of knowledge and cognition" (Harley 234). This model, in its increasing reliance upon mathematics and mathematical probability, essentially avoids or denies the objection that scientific observation and interpretation, and especially the technological gaze of its lens, produce anything other than objective, "real" knowledge. Through a mathematical detachment from the world, aided by the gaze of the lens, we see not the world itself (which includes us in its unmappable flux) but the numbers, straight lines and generalisations (which do not) that modern maps, including photographs, must employ to give the world fixed form and meaning. We find this model of cartography most impressively represented today in NASA's Mars Global Surveyor (MGS) probe, not merely as a sign of the technical success of mathematics, but also of its conceptual failure to provide the "true" representations of terrain to which a truly scientific cartography must aspire. MGS's 1998 attempt to "solve" the controversy surrounding a particularly contentious area on Mars, called Cydonia, with newer, "truer" images of the infamous Face on Mars was, contrary to popular opinion, an unsuccessful one; unsuccessful because NASA failed to remove all reasonable doubt that the Face was a natural geological formation. Not that this was particularly evident from media coverage of the image's release and reception -- the maverick researchers who comprised the protest were given a less than admirable hearing at the trial. Australian headlines reported that the Mars "romantics" (The Australian) had their Face theory "scuttled" (Courier-Mail). Professor Stanley V. McDaniel was demoted to "Mr McDaniel" in the The Australian, someone who wants NASA to continue re-imaging Cydonia to document other nearby features because "he believes [they] are further evidence of a Martian civilisation" (6). Also misrepresented was the like-minded, if slightly more adventurous researcher, Richard Hoagland, author of the underground classic, The Monuments of Mars: A City on the Edge of Forever (1987). Hoagland carelessly became (in both newspapers) Richard "Hoaglund", "leader of the movement" that believes the face is a monument left in Mars's Cydonia region by an ancient civilisation. The Australian, in line with the rest of the Earth's media, was apparently closing the door on the annoyingly persistent research into the Artificial Origin at Cydonia (AOC) hypothesis, which actually "does not claim that there is proof of artificial features on Mars, but that the probability of there being artificial features is strong enough to make new high-resolution photographs a top priority for any future mission to that planet" (McDaniel 2). Rather than confront the hypothesis itself, The Australian merely reminded us that "despite the image being 10 times better than the Viking photograph [a simplified qualification of the imaging process], it seems that some people still want to dream about ancient Martians building huge monuments to themselves" (The Australian). Dr. Mark J. Carlotto, a widely published specialist in the areas of digital image processing, pattern recognition, and computer vision, apparently still wants to dream that dream. An advocate of the AOC hypothesis, he notes that close examination of the image reveals the formation to be rough and highly eroded. Many have therefore concluded that the Face is natural. But others contend that if the Face is artificial it must certainly be very old and highly eroded. Thus the question remains as to how to distinguish an eroded artificial feature from a natural one. (Carlotto) This interestingly portentous question is one which NASA's Jet Propulsion Laboratory (JPL) and their contracted image processors, the privately owned Malin Space Science Systems (MSSS) have been somewhat reluctant to confront since the image of the Face was first captured in Viking frame 35A72. NASA's inaugural public statement on the Cydonia issue, for example, was handled ... "clumsily". As McDaniel reports: Upon the discovery of the Face in July, 1976, a Viking Project Scientist held up Viking frame 35A72, containing the Face, and announced to the assembled press corps that in a picture taken a few hours later "it all went away; it was just the way the light fell on it" -- but with a significant omission: the alleged later photograph, in which the facial features were supposed to have disappeared, was not shown for comparison [because] the statement could not possibly have been true. Frame 35A72 was taken in the early evening at approximately 6pm local time (sun angle 10 degrees). The object was in darkness a few hours later; the spacecraft, with an orbital period of about 24 hours, was no longer in a position to re-photograph the site; nor would it return to the site for many more orbits to come. Thus NASA's first official response to this strange object was an inexplicable misstatement based upon an apparent impossibility. (McDaniel 11-2) Things did not improve from there. NASA was never able to produce the elusive second photograph, yet continued to maintain that it both existed and conclusively disproved that the landform in question could possibly be artificial. In May 1993 the paradox attracted the attention of Senator John Glenn, who received (along with at least ten other members of the House and the Senate) a copy of the NASA document "Information on NASA's Re-Photographing of the Cydonia Region of Mars", which still held that the Face disappeared in the different lighting angles of a separate frame, which it did not reference. Finally in June 1993, after another inquiry by Senator Dianne Feinstein, a revised draft was issued which omits reference to the mysterious "real" photograph of the Face (McDaniel 12-3). NASA's next attempt to re-image Mars, including Cydonia, was the Observer probe, which failed to observe much at all and was declared lost in space. In conjunction with the recent mysterious failure of the Russian Phobos probe, and NASA's constant public relations blunders, conspiracy theories abounded. Was there something lurking at the threshold? The answer came in 1998 through the MGS, from which MSSS had produced a newer, clearer, "truer" image of the contentious feature which seemed to confirm Dr. Michael Malin's own earlier (and strangely self-contradictory) evaluation of 35A72: "it's simply a funny looking hill -- there is nothing unusual about it" (McDaniel 55). The MGS image certainly appeared, at least to the "naked eye" perusing the newspaper, to be just that. The wilderness had been tamed, and the coals cooled, temporarily. We had melted the witch. But exorcisms are never "final". Although NASA had apparently relegated the monstrous Face back to the realm of nature, restoring it within the parameters of conventional geology, McDaniel, Carlotto and others have maintained that NASA's conclusions were drastically premature, noting, as mentioned previously, that if the Face is artificial it must certainly be very old and, considering the Martian environment, highly eroded (Carlotto). According to their independent research, detailed analysis of the MGS image (which NASA appears not to have commissioned), does not invalidate the hypothesis that the Face may be artificial. Rather, it confirms many of the facial features recorded by the Viking, provides further evidence for the formation's high degree of lateral symmetry, and illuminates more anomalous internal detail (Carlotto). The Face on Mars, like the classical monsters of history, will not die easily. Ironically, perusal of Carlotto's dense research is an entry into the latent but undeniable plasticity of numbers, which itself is the quality of the monster that haunts modern cartographic representation. Mathematics, in almost every field of application, is finding it increasingly difficult to keep its disordering unknowns at bay. Geographer Erol Torun, for example, examined the angles formed by the facets of the two-mile long "D & M pyramid" (named after its discoverers, Vincent DiPietro and Gregory Molenaar). As Brian O'Leary writes, he subsequently found that the ratios between the five principal angles at the pyramid apex "express the universal mathematical constants of the square roots of 2,3,5,6, e, and pi ... . These constants should be known by any civilisation possessing Egyptian level technology (or greater) ... . The constants themselves are universal because they exist regardless of the number of the base being used". Regarding the other angles, Torun continued to find mathematically significant numbers "no matter how I looked at the object" (O'Leary 210). As the Cydonia controversy seems to clearly demonstrate, rather than revealing obvious, fixed truths about the world, mathematics and the observational tools they inspire require us to learn to see an approximation of the object they construct and represent as "real". We are thus compelled to draw the Other closer, but not "really", through the technological gaze of the artificial lens, a gaze that works to mask its own latent epistemic crisis. Indeed, this very compulsion inspired the growth of popular microscopy in the mid-eighteenth century, which required a new mode of seeing that could only very generously be termed "observation". Captain Basil Hall vividly recalls a meeting of the Geological Society, when a bottle was produced which was said to contain certain zoophytes. It was handed round, in the first instance, among the initiated on the foremost benches, who commented freely with one another on the forms of the animals in the fluid; but, when it came to our hands, we could discover nothing in the bottle but the most limpid fluid, -- without any trace, so far as our optics could make out, of animals dead or alive, the whole appearing absolutely transparent. The surprise of the ignorant at seeing nothing was only equal to that of the learned who saw so much to admire; nor was it till we were specifically instructed what we were to look for, and the shape, size, and general aspect of the zoophytes pointed out, that our understandings began to co-operate with our eyesight in peopling the fluid, which, up to that moment, had seemed perfectly uninhabited. The wonder then was, how we could possibly have omitted seeing objects now so palpable. (Mantell 8) Indeed, as Harley indicates, the relationship between the geographic and microscopic gaze is fundamental to the modern cartographic tradition. He cites Monmonier and Schnell's Map Appreciation (1988) as a recent example: Geography thrives on cartographic generalisation. The map is to the geographer what the microscope is to the microbiologist, for the ability to shrink the earth and generalise about it ... the microbiologist must choose a suitable objective lens, and the geographer must select a map scale appropriate to both the phenomenon in question and the "regional laboratory" in which the geographer is studying it. (in Harley 245) Importantly for this discussion, through both microscopy and cartography, "photography has also played a large role in twentieth-century ethnological representation", writes James Duncan. What better way to assert the primacy of the visual, produce a "true" representation of the place in question and establish presence than through the use of photography? But the mimetic claims of photography can also be called into question. A camera is a machine constructed to produce an image based upon artificial perspective. Only if one accepts the claims of the naturalness of Renaissance artificial perspective can we accept photography as a mimetic representation of the world. Such claims can be cast in doubt, for example, by the failure of peoples unfamiliar with photographs to be able to "read" them. (43) At the end of the day, though, it all may have more to do with down-to-earth economics than Martian "geopolitics". For many researchers, McDaniel among them, NASA's evasive treatment of the Face and surrounding features (variously labelled the Tholus, the D&M pyramid, the Fort, and the City Square) suggests a cover-up. Specifically, a cover-up of NASA's own inexplicable lack of investigation into apparently artificial structures on the Martian surface. Since the Jet Propulsion Laboratory's data imaging is contracted to a private company legally unaccountable to the public, McDaniel has confronted the somewhat disheartening possibility that financial motives may be obscuring investigation into what seem to be the most intriguing features of the Martian surface. He "does not personally believe in the conspiracy theory", but simply suggests that the Cydonia controversy may demonstrate that the financial interests of Malin Space Science Systems have assumed higher priority than the search for extraterrestrial artefacts: The contract for the Mars Observer (now MGS) involved close to 10 million dollars for Malin Space Science Systems ... . If it became clear that the probability of artificial structures on Mars is very high [or even that such a probability existed], it seems the focus of investigation would shift radically. The emphasis would fall to an accelerated manned mission to Mars. Archaeologists and perhaps biologists would assume an increasingly important role. It would be the manned mission (Johnson Space Flight Centre), not JPL, that takes the driver's seat. (McDaniel 1999) In other words, by producing results which indicated that the Cydonia region was worthy of closer attention, MSSS would jeopardise the future of its own multi-million dollar contract with NASA's Jet Propulsion Laboratory -- hardly a wise business venture, regardless of their geopolitical stance. But whatever the "true" source of the Martian controversy (it is, after all, the mythic planet of war), it seems undeniable that the Face on Mars is at once both an appropriately postmodern enigma and a genuine cartographic anomaly. For we find embodied within its monstrous form (through the lens), the message that our gaze is destined to be returned by ourselves. From the stars to the quarks, "we" seem to forever inhabit the very wilderness our technological gaze functions to both distance and draw closer; to abject. References Berland, Jody. "Mapping Space: Imaging Technologies and the Planetary Body." Technoscience and Cyberculture. Eds. Stanley Aronowitz, Barbara Martinsons, and Michael Menser. New York: Routledge, 1996. 123-37. Bull, Sandra. "Images from Mars Scuttle Face Theory." The Courier-Mail 8 April 1998. Carlotto, Mark J. "Analysis of Global Surveyor Imagery of the Face on Mars." 1998. 4 Oct. 2000 <http://www.psrw.com/~markc/Articles/MGSreport/paper.php>. ---. "New Cydonia Images -- April 2000: Preliminary Data Analysis." 2000. 4 Oct. 2000 <http://www.psrw.com/~markc/Articles/April_2000/April2000.php>. Crowley, Brian, and James J. Hurtak. The Face on Mars: Evidence of a Lost Martian Civilisation. 1986. Melbourne: Sun/Macmillan, 1989. Duncan, James. "Sites of Representation: Place, Time and the Discourse of the Other." Place/Culture/Representation. Eds. James Duncan and David Ley. London: Routledge, 1993. 39-56. Harley, J. B. "Deconstructing the Map." Writing Worlds: Discourse, Text and Metaphor in the Representation of Landscape. Eds. Trevor J. Barnes and James Duncan. London: Routledge, 1992. 231-47. Leech, Graeme. "Mars Romantics Face the Truth: There's Nothing Out There." The Australian 8 April 1998. Mantell, Gideon Algernon. The Invisible World Revealed by the Microscope; or, Thoughts on Animalcules. London: John Murray, 1850. McDaniel, Stanley V. The McDaniel Report: On the Failure of Executive, Congressional and Scientific Responsibility in Investigating Possible Evidence of Artificial Structures on the Surface of Mars and in Setting Mission Priorities for NASA's Mars Exploration Program. Berkeley: North Atlantic, 1993. ---. "Here It Is! But What Is It?" 1998. 4 Oct. 2000 <http://www.mcdanielreport.com/homepage.htm>. ---. "The Cydonia Question: Where Do We Stand?" 1999. 4 Oct. 2000. <http://www.mcdanielreport.com/standing.php>. O'Leary, Brian. "Mars and the Search for Extraterrestrial Life." Suppressed Inventions and Other Discoveries. Ed. Jonathan Eisen. Auckland: AIT, 1994. 204-13. Picknett, Lynn, and Clive Prince. The Stargate Conspiracy: Revealing the Truth behind Extraterrestrial Contact, Military Intelligence and the Mysteries of Ancient Egypt. London: Little, Brown and Co., 1999. Citation reference for this article MLA style: Adam Dodd. "'Unacceptable Renewals': The Geopolitics of Martian Cartography." M/C: A Journal of Media and Culture 3.6 (2000). [your date of access] <http://www.api-network.com/mc/0012/mars.php>. Chicago style: Adam Dodd, "'Unacceptable Renewals': The Geopolitics of Martian Cartography," M/C: A Journal of Media and Culture 3, no. 6 (2000), <http://www.api-network.com/mc/0012/mars.php> ([your date of access]). APA style: Adam Dodd. (2000) 'Unacceptable renewals': the geopolitics of Martian cartography. M/C: A Journal of Media and Culture 3(6). <http://www.api-network.com/mc/0012/mars.php> ([your date of access]).

Introduction This article is a study in anxiety with regard to social online spaces (SOS) conceived of as dark. There are two possible ways to define ‘dark’ in this context. The first is that communication is dark because it either has limited distribution, is not open to all users (closed groups are a case example) or hidden. The second definition, linked as a result of the first, is the way that communication via these means is interpreted and understood. Dark social spaces disrupt the accepted top-down flow by the ‘gazing elite’ (data aggregators including social media), but anxious users might need to strain to notice what is out there, and this in turn destabilises one’s reception of the scene. In an environment where surveillance technologies are proliferating, this article examines contemporary, dark, interconnected, and interactive communications for the entangled affordances that might be brought to bear. A provocation is that resistance through counterveillance or “sousveillance” is one possibility. An alternative (or addition) is retreating to or building ‘dark’ spaces that are less surveilled and (perhaps counterintuitively) less fearful. This article considers critically the notion of dark social online spaces via four broad socio-technical concerns connected to the big social media services that have helped increase a tendency for fearful anxiety produced by surveillance and the perceived implications for personal privacy. It also shines light on the aspect of darkness where some users are spurred to actively seek alternative, dark social online spaces. Since the 1970s, public-key cryptosystems typically preserved security for websites, emails, and sensitive health, government, and military data, but this is now reduced (Williams). We have seen such systems exploited via cyberattacks and misappropriated data acquired by affiliations such as Facebook-Cambridge Analytica for targeted political advertising during the 2016 US elections. Via the notion of “parasitic strategies”, such events can be described as news/information hacks “whose attack vectors target a system’s weak points with the help of specific strategies” (von Nordheim and Kleinen-von Königslöw, 88). In accord with Wilson and Serisier’s arguments (178), emerging technologies facilitate rapid data sharing, collection, storage, and processing wherein subsequent “outcomes are unpredictable”. This would also include the effect of acquiescence. In regard to our digital devices, for some, being watched overtly—through cameras encased in toys, computers, and closed-circuit television (CCTV) to digital street ads that determine the resonance of human emotions in public places including bus stops, malls, and train stations—is becoming normalised (McStay, Emotional AI). It might appear that consumers immersed within this Internet of Things (IoT) are themselves comfortable interacting with devices that record sound and capture images for easy analysis and distribution across the communications networks. A counter-claim is that mainstream social media corporations have cultivated a sense of digital resignation “produced when people desire to control the information digital entities have about them but feel unable to do so” (Draper and Turow, 1824). Careful consumers’ trust in mainstream media is waning, with readers observing a strong presence of big media players in the industry and are carefully picking their publications and public intellectuals to follow (Mahmood, 6). A number now also avoid the mainstream internet in favour of alternate dark sites. This is done by users with “varying backgrounds, motivations and participation behaviours that may be idiosyncratic (as they are rooted in the respective person’s biography and circumstance)” (Quandt, 42). By way of connection with dark internet studies via Biddle et al. (1; see also Lasica), the “darknet” is a collection of networks and technologies used to share digital content … not a separate physical network but an application and protocol layer riding on existing networks. Examples of darknets are peer-to-peer file sharing, CD and DVD copying, and key or password sharing on email and newsgroups. As we note from the quote above, the “dark web” uses existing public and private networks that facilitate communication via the Internet. Gehl (1220; see also Gehl and McKelvey) has detailed that this includes “hidden sites that end in ‘.onion’ or ‘.i2p’ or other Top-Level Domain names only available through modified browsers or special software. Accessing I2P sites requires a special routing program ... . Accessing .onion sites requires Tor [The Onion Router]”. For some, this gives rise to social anxiety, read here as stemming from that which is not known, and an exaggerated sense of danger, which makes fight or flight seem the only options. This is often justified or exacerbated by the changing media and communication landscape and depicted in popular documentaries such as The Social Dilemma or The Great Hack, which affect public opinion on the unknown aspects of internet spaces and the uses of personal data. The question for this article remains whether the fear of the dark is justified. Consider that most often one will choose to make one’s intimate bedroom space dark in order to have a good night’s rest. We might pleasurably escape into a cinema’s darkness for the stories told therein, or walk along a beach at night enjoying unseen breezes. Most do not avoid these experiences, choosing to actively seek them out. Drawing this thread, then, is the case made here that agency can also be found in the dark by resisting socio-political structural harms. 1. Digital Futures and Anxiety of the Dark Fear of the darkI have a constant fear that something's always nearFear of the darkFear of the darkI have a phobia that someone's always there In the lyrics to the song “Fear of the Dark” (1992) by British heavy metal group Iron Maiden is a sense that that which is unknown and unseen causes fear and anxiety. Holding a fear of the dark is not unusual and varies in degree for adults as it does for children (Fellous and Arbib). Such anxiety connected to the dark does not always concern darkness itself. It can also be a concern for the possible or imagined dangers that are concealed by the darkness itself as a result of cognitive-emotional interactions (McDonald, 16). Extending this claim is this article’s non-binary assertion that while for some technology and what it can do is frequently misunderstood and shunned as a result, for others who embrace the possibilities and actively take it on it is learning by attentively partaking. Mistakes, solecism, and frustrations are part of the process. Such conceptual theorising falls along a continuum of thinking. Global interconnectivity of communications networks has certainly led to consequent concerns (Turkle Alone Together). Much focus for anxiety has been on the impact upon social and individual inner lives, levels of media concentration, and power over and commercialisation of the internet. Of specific note is that increasing commercial media influence—such as Facebook and its acquisition of WhatsApp, Oculus VR, Instagram, CRTL-labs (translating movements and neural impulses into digital signals), LiveRail (video advertising technology), Chainspace (Blockchain)—regularly changes the overall dynamics of the online environment (Turow and Kavanaugh). This provocation was born out recently when Facebook disrupted the delivery of news to Australian audiences via its service. Mainstream social online spaces (SOS) are platforms which provide more than the delivery of media alone and have been conceptualised predominantly in a binary light. On the one hand, they can be depicted as tools for the common good of society through notional widespread access and as places for civic participation and discussion, identity expression, education, and community formation (Turkle; Bruns; Cinque and Brown; Jenkins). This end of the continuum of thinking about SOS seems set hard against the view that SOS are operating as businesses with strategies that manipulate consumers to generate revenue through advertising, data, venture capital for advanced research and development, and company profit, on the other hand. In between the two polar ends of this continuum are the range of other possibilities, the shades of grey, that add contemporary nuance to understanding SOS in regard to what they facilitate, what the various implications might be, and for whom. By way of a brief summary, anxiety of the dark is steeped in the practices of privacy-invasive social media giants such as Facebook and its ancillary companies. Second are the advertising technology companies, surveillance contractors, and intelligence agencies that collect and monitor our actions and related data; as well as the increased ease of use and interoperability brought about by Web 2.0 that has seen a disconnection between technological infrastructure and social connection that acts to limit user permissions and online affordances. Third are concerns for the negative effects associated with depressed mental health and wellbeing caused by “psychologically damaging social networks”, through sleep loss, anxiety, poor body image, real world relationships, and the fear of missing out (FOMO; Royal Society for Public Health (UK) and the Young Health Movement). Here the harms are both individual and societal. Fourth is the intended acceleration toward post-quantum IoT (Fernández-Caramés), as quantum computing’s digital components are continually being miniaturised. This is coupled with advances in electrical battery capacity and interconnected telecommunications infrastructures. The result of such is that the ontogenetic capacity of the powerfully advanced network/s affords supralevel surveillance. What this means is that through devices and the services that they provide, individuals’ data is commodified (Neff and Nafus; Nissenbaum and Patterson). Personal data is enmeshed in ‘things’ requiring that the decisions that are both overt, subtle, and/or hidden (dark) are scrutinised for the various ways they shape social norms and create consequences for public discourse, cultural production, and the fabric of society (Gillespie). Data and personal information are retrievable from devices, sharable in SOS, and potentially exposed across networks. For these reasons, some have chosen to go dark by being “off the grid”, judiciously selecting their means of communications and their ‘friends’ carefully. 2. Is There Room for Privacy Any More When Everyone in SOS Is Watching? An interesting turn comes through counterarguments against overarching institutional surveillance that underscore the uses of technologies to watch the watchers. This involves a practice of counter-surveillance whereby technologies are tools of resistance to go ‘dark’ and are used by political activists in protest situations for both communication and avoiding surveillance. This is not new and has long existed in an increasingly dispersed media landscape (Cinque, Changing Media Landscapes). For example, counter-surveillance video footage has been accessed and made available via live-streaming channels, with commentary in SOS augmenting networking possibilities for niche interest groups or micropublics (Wilson and Serisier, 178). A further example is the Wordpress site Fitwatch, appealing for an end to what the site claims are issues associated with police surveillance (fitwatch.org.uk and endpolicesurveillance.wordpress.com). Users of these sites are called to post police officers’ identity numbers and photographs in an attempt to identify “cops” that might act to “misuse” UK Anti-terrorism legislation against activists during legitimate protests. Others that might be interested in doing their own “monitoring” are invited to reach out to identified personal email addresses or other private (dark) messaging software and application services such as Telegram (freeware and cross-platform). In their work on surveillance, Mann and Ferenbok (18) propose that there is an increase in “complex constructs between power and the practices of seeing, looking, and watching/sensing in a networked culture mediated by mobile/portable/wearable computing devices and technologies”. By way of critical definition, Mann and Ferenbok (25) clarify that “where the viewer is in a position of power over the subject, this is considered surveillance, but where the viewer is in a lower position of power, this is considered sousveillance”. It is the aspect of sousveillance that is empowering to those using dark SOS. One might consider that not all surveillance is “bad” nor institutionalised. It is neither overtly nor formally regulated—as yet. Like most technologies, many of the surveillant technologies are value-neutral until applied towards specific uses, according to Mann and Ferenbok (18). But this is part of the ‘grey area’ for understanding the impact of dark SOS in regard to which actors or what nations are developing tools for surveillance, where access and control lies, and with what effects into the future. 3. Big Brother Watches, So What Are the Alternatives: Whither the Gazing Elite in Dark SOS? By way of conceptual genealogy, consideration of contemporary perceptions of surveillance in a visually networked society (Cinque, Changing Media Landscapes) might be usefully explored through a revisitation of Jeremy Bentham’s panopticon, applied here as a metaphor for contemporary surveillance. Arguably, this is a foundational theoretical model for integrated methods of social control (Foucault, Surveiller et Punir, 192-211), realised in the “panopticon” (prison) in 1787 by Jeremy Bentham (Bentham and Božovič, 29-95) during a period of social reformation aimed at the improvement of the individual. Like the power for social control over the incarcerated in a panopticon, police power, in order that it be effectively exercised, “had to be given the instrument of permanent, exhaustive, omnipresent surveillance, capable of making all visible … like a faceless gaze that transformed the whole social body into a field of perception” (Foucault, Surveiller et Punir, 213–4). In grappling with the impact of SOS for the individual and the collective in post-digital times, we can trace out these early ruminations on the complex documentary organisation through state-controlled apparatuses (such as inspectors and paid observers including “secret agents”) via Foucault (Surveiller et Punir, 214; Subject and Power, 326-7) for comparison to commercial operators like Facebook. Today, artificial intelligence (AI), facial recognition technology (FRT), and closed-circuit television (CCTV) for video surveillance are used for social control of appropriate behaviours. Exemplified by governments and the private sector is the use of combined technologies to maintain social order, from ensuring citizens cross the street only on green lights, to putting rubbish in the correct recycling bin or be publicly shamed, to making cashless payments in stores. The actions see advantages for individual and collective safety, sustainability, and convenience, but also register forms of behaviour and attitudes with predictive capacities. This gives rise to suspicions about a permanent account of individuals’ behaviour over time. Returning to Foucault (Surveiller et Punir, 135), the impact of this finds a dissociation of power from the individual, whereby they become unwittingly impelled into pre-existing social structures, leading to a ‘normalisation’ and acceptance of such systems. If we are talking about the dark, anxiety is key for a Ministry of SOS. Following Foucault again (Subject and Power, 326-7), there is the potential for a crawling, creeping governance that was once distinct but is itself increasingly hidden and growing. A blanket call for some form of ongoing scrutiny of such proliferating powers might be warranted, but with it comes regulation that, while offering certain rights and protections, is not without consequences. For their part, a number of SOS platforms had little to no moderation for explicit content prior to December 2018, and in terms of power, notwithstanding important anxiety connected to arguments that children and the vulnerable need protections from those that would seek to take advantage, this was a crucial aspect of community building and self-expression that resulted in this freedom of expression. In unearthing the extent that individuals are empowered arising from the capacity to post sexual self-images, Tiidenberg ("Bringing Sexy Back") considered that through dark SOS (read here as unregulated) some users could work in opposition to the mainstream consumer culture that provides select and limited representations of bodies and their sexualities. This links directly to Mondin’s exploration of the abundance of queer and feminist pornography on dark SOS as a “counterpolitics of visibility” (288). This work resulted in a reasoned claim that the technological structure of dark SOS created a highly political and affective social space that users valued. What also needs to be underscored is that many users also believed that such a space could not be replicated on other mainstream SOS because of the differences in architecture and social norms. Cho (47) worked with this theory to claim that dark SOS are modern-day examples in a history of queer individuals having to rely on “underground economies of expression and relation”. Discussions such as these complicate what dark SOS might now become in the face of ‘adult’ content moderation and emerging tracking technologies to close sites or locate individuals that transgress social norms. Further, broader questions are raised about how content moderation fits in with the public space conceptualisations of SOS more generally. Increasingly, “there is an app for that” where being able to identify the poster of an image or an author of an unknown text is seen as crucial. While there is presently no standard approach, models for combining instance-based and profile-based features such as SVM for determining authorship attribution are in development, with the result that potentially far less content will remain hidden in the future (Bacciu et al.). 4. There’s Nothing New under the Sun (Ecclesiastes 1:9) For some, “[the] high hopes regarding the positive impact of the Internet and digital participation in civic society have faded” (Schwarzenegger, 99). My participant observation over some years in various SOS, however, finds that critical concern has always existed. Views move along the spectrum of thinking from deep scepticisms (Stoll, Silicon Snake Oil) to wondrous techo-utopian promises (Negroponte, Being Digital). Indeed, concerns about the (then) new technologies of wireless broadcasting can be compared with today’s anxiety over the possible effects of the internet and SOS. Inglis (7) recalls, here, too, were fears that humanity was tampering with some dangerous force; might wireless wave be causing thunderstorms, droughts, floods? Sterility or strokes? Such anxieties soon evaporated; but a sense of mystery might stay longer with evangelists for broadcasting than with a laity who soon took wireless for granted and settled down to enjoy the products of a process they need not understand. As the analogy above makes clear, just as audiences came to use ‘the wireless’ and later the internet regularly, it is reasonable to argue that dark SOS will also gain widespread understanding and find greater acceptance. Dark social spaces are simply the recent development of internet connectivity and communication more broadly. The dark SOS afford choice to be connected beyond mainstream offerings, which some users avoid for their perceived manipulation of content and user both. As part of the wider array of dark web services, the resilience of dark social spaces is reinforced by the proliferation of users as opposed to decentralised replication. Virtual Private Networks (VPNs) can be used for anonymity in parallel to TOR access, but they guarantee only anonymity to the client. A VPN cannot guarantee anonymity to the server or the internet service provider (ISP). While users may use pseudonyms rather than actual names as seen on Facebook and other SOS, users continue to take to the virtual spaces they inhabit their off-line, ‘real’ foibles, problems, and idiosyncrasies (Chenault). To varying degrees, however, people also take their best intentions to their interactions in the dark. The hyper-efficient tools now deployed can intensify this, which is the great advantage attracting some users. In balance, however, in regard to online information access and dissemination, critical examination of what is in the public’s interest, and whether content should be regulated or controlled versus allowing a free flow of information where users self-regulate their online behaviour, is fraught. O’Loughlin (604) was one of the first to claim that there will be voluntary loss through negative liberty or freedom from (freedom from unwanted information or influence) and an increase in positive liberty or freedom to (freedom to read or say anything); hence, freedom from surveillance and interference is a kind of negative liberty, consistent with both libertarianism and liberalism. Conclusion The early adopters of initial iterations of SOS were hopeful and liberal (utopian) in their beliefs about universality and ‘free’ spaces of open communication between like-minded others. This was a way of virtual networking using a visual motivation (led by images, text, and sounds) for consequent interaction with others (Cinque, Visual Networking). The structural transformation of the public sphere in a Habermasian sense—and now found in SOS and their darker, hidden or closed social spaces that might ensure a counterbalance to the power of those with influence—towards all having equal access to platforms for presenting their views, and doing so respectfully, is as ever problematised. Broadly, this is no more so, however, than for mainstream SOS or for communicating in the world. References Bacciu, Andrea, Massimo La Morgia, Alessandro Mei, Eugenio Nerio Nemmi, Valerio Neri, and Julinda Stefa. “Cross-Domain Authorship Attribution Combining Instance Based and Profile-Based Features.” CLEF (Working Notes). Lugano, Switzerland, 9-12 Sep. 2019. Bentham, Jeremy, and Miran Božovič. The Panopticon Writings. London: Verso Trade, 1995. Biddle, Peter, et al. “The Darknet and the Future of Content Distribution.” Proceedings of the 2002 ACM Workshop on Digital Rights Management. Vol. 6. Washington DC, 2002. Bruns, Axel. Blogs, Wikipedia, Second Life, and Beyond: From Production to Produsage. New York: Peter Lang, 2008. Chenault, Brittney G. “Developing Personal and Emotional Relationships via Computer-Mediated Communication.” CMC Magazine 5.5 (1998). 1 May 2020 <http://www.december.com/cmc/mag/1998/may/chenault.html>. Cho, Alexander. “Queer Reverb: Tumblr, Affect, Time.” Networked Affect. Eds. K. Hillis, S. Paasonen, and M. Petit. Cambridge, Mass.: MIT Press, 2015: 43-58. Cinque, Toija. Changing Media Landscapes: Visual Networking. London: Oxford UP, 2015. ———. “Visual Networking: Australia's Media Landscape.” Global Media Journal: Australian Edition 6.1 (2012): 1-8. Cinque, Toija, and Adam Brown. “Educating Generation Next: Screen Media Use, Digital Competencies, and Tertiary Education.” Digital Culture & Education 7.1 (2015). Draper, Nora A., and Joseph Turow. “The Corporate Cultivation of Digital Resignation.” New Media & Society 21.8 (2019): 1824-1839. Fellous, Jean-Marc, and Michael A. Arbib, eds. Who Needs Emotions? The Brain Meets the Robot. New York: Oxford UP, 2005. Fernández-Caramés, Tiago M. “From Pre-Quantum to Post-Quantum IoT Security: A Survey on Quantum-Resistant Cryptosystems for the Internet of Things.” IEEE Internet of Things Journal 7.7 (2019): 6457-6480. Foucault, Michel. Surveiller et Punir: Naissance de la Prison [Discipline and Punish—The Birth of The Prison]. Trans. Alan Sheridan. New York: Random House, 1977. Foucault, Michel. “The Subject and Power.” Michel Foucault: Power, the Essential Works of Michel Foucault 1954–1984. Vol. 3. Trans. R. Hurley and others. Ed. J.D. Faubion. London: Penguin, 2001. Gehl, Robert W. Weaving the Dark Web: Legitimacy on Freenet, Tor, and I2P. Cambridge, Massachusetts: MIT Press, 2018. Gehl, Robert, and Fenwick McKelvey. “Bugging Out: Darknets as Parasites of Large-Scale Media Objects.” Media, Culture & Society 41.2 (2019): 219-235. Gillespie, Tarleton. Custodians of the Internet: Platforms, Content Moderation, and the Hidden Decisions That Shape Social Media. London: Yale UP, 2018. Habermas, Jürgen. The Structural Transformation of the Public Sphere: An Inquiry into a Category of Bourgeois Society. Trans. Thomas Burger with the assistance of Frederick Lawrence. Cambridge, Mass.: MIT Press, 1989. Inglis, Ken S. This Is the ABC: The Australian Broadcasting Commission 1932–1983. Melbourne: Melbourne UP, 1983. Iron Maiden. “Fear of the Dark.” London: EMI, 1992. Jenkins, Henry. Convergence Culture: Where Old and New Media Collide. New York: New York UP, 2006. Lasica, J. D. Darknet: Hollywood’s War against the Digital Generation. New York: John Wiley and Sons, 2005. Mahmood, Mimrah. “Australia's Evolving Media Landscape.” 13 Apr. 2021 <https://www.meltwater.com/en/resources/australias-evolving-media-landscape>. Mann, Steve, and Joseph Ferenbok. “New Media and the Power Politics of Sousveillance in a Surveillance-Dominated World.” Surveillance & Society 11.1/2 (2013): 18-34. McDonald, Alexander J. “Cortical Pathways to the Mammalian Amygdala.” Progress in Neurobiology 55.3 (1998): 257-332. McStay, Andrew. Emotional AI: The Rise of Empathic Media. London: Sage, 2018. Mondin, Alessandra. “‘Tumblr Mostly, Great Empowering Images’: Blogging, Reblogging and Scrolling Feminist, Queer and BDSM Desires.” Journal of Gender Studies 26.3 (2017): 282-292. Neff, Gina, and Dawn Nafus. Self-Tracking. Cambridge, Mass.: MIT Press, 2016. Negroponte, Nicholas. Being Digital. New York: Alfred A. Knopf, 1995. Nissenbaum, Helen, and Heather Patterson. “Biosensing in Context: Health Privacy in a Connected World.” Quantified: Biosensing Technologies in Everyday Life. Ed. Dawn Nafus. 2016. 68-79. O’Loughlin, Ben. “The Political Implications of Digital Innovations.” Information, Communication and Society 4.4 (2001): 595–614. Quandt, Thorsten. “Dark Participation.” Media and Communication 6.4 (2018): 36-48. Royal Society for Public Health (UK) and the Young Health Movement. “#Statusofmind.” 2017. 2 Apr. 2021 <https://www.rsph.org.uk/our-work/campaigns/status-of-mind.html>. Statista. “Number of IoT devices 2015-2025.” 27 Nov. 2020 <https://www.statista.com/statistics/471264/iot-number-of-connected-devices-worldwide/>. Schwarzenegger, Christian. “Communities of Darkness? Users and Uses of Anti-System Alternative Media between Audience and Community.” Media and Communication 9.1 (2021): 99-109. Stoll, Clifford. Silicon Snake Oil: Second Thoughts on the Information Highway. Anchor, 1995. Tiidenberg, Katrin. “Bringing Sexy Back: Reclaiming the Body Aesthetic via Self-Shooting.” Cyberpsychology: Journal of Psychosocial Research on Cyberspace 8.1 (2014). The Great Hack. Dirs. Karim Amer, Jehane Noujaim. Netflix, 2019. The Social Dilemma. Dir. Jeff Orlowski. Netflix, 2020. Turkle, Sherry. The Second Self: Computers and the Human Spirit. Cambridge, Mass.: MIT Press, 2005. Turkle, Sherry. Alone Together: Why We Expect More from Technology and Less from Each Other. UK: Hachette, 2017. Turow, Joseph, and Andrea L. Kavanaugh, eds. The Wired Homestead: An MIT Press Sourcebook on the Internet and the Family. Cambridge, Mass.: MIT Press, 2003. Von Nordheim, Gerret, and Katharina Kleinen-von Königslöw. “Uninvited Dinner Guests: A Theoretical Perspective on the Antagonists of Journalism Based on Serres’ Parasite.” Media and Communication 9.1 (2021): 88-98. Williams, Chris K. “Configuring Enterprise Public Key Infrastructures to Permit Integrated Deployment of Signature, Encryption and Access Control Systems.” MILCOM 2005-2005 IEEE Military Communications Conference. IEEE, 2005. Wilson, Dean, and Tanya Serisier. “Video Activism and the Ambiguities of Counter-Surveillance.” Surveillance & Society 8.2 (2010): 166-180.

Il seguente progetto è volto ad implementare sul microcontroller "Raspberry Pi v.4 Model B" un software utilizzante a mo' di scatola nera, anche, in parte, una rete neurale che sulla base di una classificazione precedentemente realizzata da soggetti terzi e sulla base di opportuni modelli preaddestrati sfrutti un meccanismo di apprendimento supervisionato per stimare ragionevolmente, secondo opportuni criteri, il sesso, la fascia d'età, lo stato emotivo (caricaturale, ossia forzato) e la distanza approssimativa di uno o più soggetti ripresi frontalmente in volto da una telecamera in tempo reale.

Le fotocamere digitali acquisiscono, manipolano e salvano in memoria informazioni elettronicamente tramite un sensore e unitá di calcolo integrate, rendendo la fotografia accessibile agli utenti finali i quali non necessitano di basi di conoscenza di chimica o delle specifiche operazioni meccaniche che erano necessarie per lo sviluppo con la fotografia analogica. Diversi tipi di degradazioni e artefatti possono affliggere le immagini acquisite tramite camere digitali, riducendo la fedeltà percepita delle immagini e rendendo piú difficili operazioni di visione artificiale che possono essere operate sulle immagini acquisite. Tre elementi possono essere identificati come possibili sorgenti di artefatti in un'immagine: il contenuto della scena, le limitazioni hardware della camera e la pipeline di elaborazione dell’immagine, dall’acquisizione al salvataggio in memoria. Alcuni artefatti non sono direttamente trattati all’interno della tipica pipeline di camera, come ad esempio la presenza di nebbia o pioggia che possono ridurre la visibilitá della scena acquisita. Questi artefatti , per essere trattati, necessitano di metodologie ad hoc che sono generalmente applicate come operazioni post-processing. Altri tipi di artefatti sono legati ai processi di elaborazione dell’immagine e all'intera pipeline di camera integrata nelle camere. Questi artefatti includono il rumore proveniente dal sensore, dominanti cromatiche indesiderate, basso contrasto e artefatti di compressione. L’obiettivo di questa tesi è l'identificazione e la progettazione di nuovi e piú robusti moduli di elaborazione e restauro dell’immagine capaci di migliorare la qualitá delle immagini acquisite dalle camere digitali, in particolare in scenari critici quali condizioni climatiche avverse, condizione di bassa illuminazione etc… Gli artefatti identificati sono divisi in due macro gruppi: “artefatti in-camera” e “artefatti esterni” Nel primo gruppo sono stati identificati e trattati quattro argomenti: la rimozione del rumore proveniente dal sensore della camera, il processo di bilanciamento automatico del bianco, il miglioramento automatico del contrasto e la rimozione di artefatti di compressione JPEG. Il processo di progettazione di questi moduli ha tenuto conto di aspetti di efficienza, in termini di calcolo e memoria occupata, in relazione alla possibilità di integrare tali moduli in future configurazioni di pipeline di camera. Il secondo gruppo di artefatti è legato alla presenza di elementi nella scena che possono causare una degradazione in termini di fedeltà dell’immagine e/o usabilità. In particolare il lavoro presentato nella tesi è focalizzato su artefatti provenienti dalla presenza di pioggia nella scena scattata. La tesi, dopo una descrizione della tipica pipeline di elaborazione delle camere digitali, analizza i diversi tipi di artefatti che possono affliggere la qualità di una immagine, e descrive la progettazione delle soluzioni proposte. Tutti gli approcci proposti sono basati su tecniche di apprendimento automatico, come ad esempio Reti Neurali Convoluzionali o procedure di ottimizzazione Bayesiana, e sono stati validati sperimentalmente su dataset standard. I contributi principali di questa tesi possono essere riassunti in tre punti: integrazione di approcci classici di imaging con tecniche di ottimizzazione basate su machine learning, progettazione e sviluppo di nuovi approcci e architetture di deep learning per il restauro e l'elaborazione di immagini e analisi di metodi di image processing basati su deep learning in task di computer vision.
Digital cameras record, manipulate, and store information electronically through sensors and built-in computers, which makes photography more available to final users which do not anymore need to rely on the use of chemicals and knowledge of mechanical procedures to develop their pictures. Different types of degradation and artifacts can affect images acquired using digital cameras, decreasing the perceptual fidelity of images and making harder many image processing and analysis tasks that can be performed on the collected images. Three elements can be identified as possible sources of artifacts in an image: the scene content, the hardware limitations and flaws, and finally the operations performed by the digital camera processing pipeline itself, from acquisition to compression and storing. Some artifacts are not directly treated in the typical camera processing pipeline, such as the presence of haze or rain that can reduce visibility of the scene in the depicted images. These artifacts require the design of ad hoc methods that are usually applied as post-processing on the acquired images. Other types of artifacts are related to the imaging process and to the image processing pipeline implemented on board of digital cameras. These include sensor noise, undesirable color cast, poor contrast and compression artifacts. The objective of this thesis is the identification and design of new and more robust modules for image processing and restoration that can improve the quality of the acquired images, in particular in critical scenarios such as adverse weather conditions, poor light in the scene etc… . The artifacts identified are divided into two main groups: “in camera-generated artifacts" and “external artifacts and problems". In the first group it has been identified and addressed four main issues: sensor camera noise removal, automatic white balancing, automatic contrast enhancement and compression artifacts removal. The design process of the proposed solutions has considered efficiency aspects, due to the possibility of directly integrating them in future camera pipelines. The second group of artifacts are related to the presence of elements in the scene which may cause a degradation in terms of visual fidelity and/or usability of the images. In particular the focus is on artifacts induced by the presence of rain in the scene. The thesis, after a brief review of the digital camera processing pipeline, analyzes the different types of artifacts that can affect image quality, and describes the design of the proposed solutions. All the proposed approaches are based on machine learning techniques, such as Convolutional Neural Networks and Bayesian optimization procedure, and are experimentally validated on standard images datasets. The overall contributions of this thesis can be summarized in three points: integration of classical imaging approaches with machine learning optimization techniques, design of novel deep learning architectures and approaches and analysis and application of deep learning image processing algorithms in other computer vision tasks.

Attualmente la Computer Vision, disciplina che consente di estrarre informazioni a partire da immagini digitali, è uno dei settori informatici più in fermento. Grazie alle recenti conquiste e progressi, tale settore ha raggiunto uno stato di maturità tale da poter essere applicato in svariati ambiti, a partire da quello industriale, fino ad arrivare ad applicazioni più vicine alla vita quotidiana. In particolare, si è raggiunto uno stato dell'arte sempre più solido nel campo del riconoscimento di oggetti (object detection) grazie allo sviluppo delle Convolutional Neural Networks (CNN): sistemi che si basano su un modello matematico, che viene gradualmente raffinato in base all'esperienza stessa del sistema nell'esecuzione di questo task, acquisita mediante tecniche di machine learning. Grazie a ciò, le CNN sono in grado di riconoscere e classificare il contenuto di immagini, dando loro una semantica. Tali sistemi però richiedono una grande capacità computazionale ed un'ingente quantità di memoria, pertanto la loro esecuzione avviene maggiormente su architetture potenti, come le GPU. Nonostante ciò, una delle sfide attualmente più importanti riguarda la classificazione in tempo reale di immagini eseguendo le reti neurali convolutive anche su architetture con disponibilità energetica e capacità computazionali ridotte, quali sono i sistemi embedded. Quindi, nel seguente trattato si propone un'implementazione di CNN riconfigurabile realizzata in linguaggio C. Ciò è risultato in un sistema semplice e modulare che con diverse ottimizzazioni ad-hoc può essere considerato un buon candidato per il porting su architetture embedded riconfigurabili FPGA.

La Normalizzazione del Mercato Europeo e Mondiale, unita al rinnovamento degli sforzi verso una innovazione tecnologica “globale”, ha contribuito allo sviluppo di un “Sistema di Qualità” che impone alla produzione industriale Nazionale la necessità di sfruttare tecnologie produttive innovative al fine di consentire una sempre maggiore flessibilità ed automatizzazione permettendo altresì un mantenimento dell’attuale status di leader europeo del settore. La tendenza verso una produzione che veda minimizzate le eventuali difettosità di un prodotto, intese anche come rispondenza del prodotto a specifiche esigenze, e per la quale siano controllabili in tempo reale i parametri critici del flusso produttivo, costituisce uno degli aspetti maggiormente rilevanti nell’ambito del miglioramento qualitativo dei prodotti per le imprese del territorio. Un accurato sistema di controllo della qualità del prodotto, che si affianchi ed integri il controllo del processo produttivo, diviene dunque condizione indispensabile sia per la migliore rispondenza dei prodotti alle aspettative della clientela, sia per il miglioramento della qualità del prodotto finito, per l’aumento della produzione e per la diminuzione dei costi. La diversificazione delle produzioni tradizionali ha comportato, inoltre, una parallela riqualificazione del tessuto produttivo per cui alcune aziende del settore, ed in particolare nel distretto tessile di Prato, hanno acquisito un ruolo più consapevole potenziando sia la progettazione di nuovi prodotti che le tecniche di controllo qualità. In questo complesso ed articolato contesto assumono importanza crescente i sistemi di controllo, misura o caratterizzazione del prodotto basati su tecniche di Visione Artificiale e su simulazione e controllo dei processi industriali mediante utilizzo di tecniche di Intelligenza Artificiale (IA). Tali sistemi sono correntemente utilizzati dall’industria dell’automobile, meccanica, farmaceutica, alimentare, del confezionamento e dell’assemblaggio allo scopo sia di automatizzare l’ispezione industriale dei prodotti che di regolare ed ottimizzare i parametri chiave dei processi produttivi. Ne sono esempi tipici il controllo dimensionale, il rilevamento della posizione o l’analisi dello stato superficiale di oggetti o utensili, l’individuazione di difetti e la verifica del corretto montaggio. Inoltre tali tecniche costituiscono uno strumento in grado di acquisire ed analizzare le caratteristiche quantitative dei prodotti al fine di ottimizzare la linea produttiva. Per tali ragioni il Dipartimento di Meccanica e Tecnologie Industriali (DMTI) dell’Università degli Studi di Firenze ha inteso, con il presente lavoro di Dottorato, “esportare” le proprie competenze e la propria esperienza scientifica ed applicativa sulla progettazione, e successiva implementazione, di sistemi basati sulla Visione Artificiale e sulla Intelligenza Artificiale in grado di compiere operazioni complesse su prodotti tessili. In particolare due delle tecniche più comunemente impiegate per assolvere i compiti descritti sono quelle basate sul processamento digitale delle immagini e sulle reti neurali artificiali; spesso tali tecniche sono unite tra loro per formare i cosiddetti sistemi di visione intelligente. Nel presente lavoro, dunque, saranno presentati tre diversi casi di studio delle suddette tecniche applicate a prodotti e processi della filiera tessile. In particolare saranno presentate le problematiche riscontrate, i vincoli progettuali, le esigenze di innovazione, le metodologie impiegate al fine di rispettare i requisiti imposti, le azioni di ricerca ed infine i risultati conseguiti. I tre casi, che hanno previsto un lavoro di tre anni e sono stati svolti in collaborazione con il centro di Ricerca Tessile Tecnotessile s.r.l., sono presentati nel testo secondo un criterio di complessità crescente. Nel Capitolo 1 si riporta una rapida descrizione delle attuali tecniche di misura fisica su tessuti; nel Capitolo 2 sono presentate le principali tecniche di Visione Artificiale e di trattamento delle immagini digitali (Image Processing); nel Capitolo 3 sono descritte accuratamente le tipologie di Reti Neurali utilizzate per lo sviluppo delle tematiche descritte. Il primo caso di studio (Capitolo 4) riguarda lo sviluppo e la successiva implementazione di un sistema (prototipo) in grado di compiere in automatico una classificazione di tessuti da riciclare (prevalentemente lanieri) sulla base della risposta spettrofotometrica degli stessi (colore e proprietà di riflessione/assorbimento). L’obiettivo (specifica di progetto) è quello di classificare i prodotti con le stesse modalità con cui viene effettuata la selezione da esperti (cernitori), con una precisione paragonabile e con maggiore ripetibilità. Allo scopo è stato implementato un sistema basato sulla combinazione tra tecniche statistiche e Reti Neurali in grado di classificare i tessuti con una accuratezza del 93.5 %, in modo automatico, non intrusivo e caratterizzato dalla massima ripetibilità. Il secondo caso di studio (Capitolo 5) è relativo allo sviluppo di un prototipo di sistema di Visione Artificiale in grado di individuare e classificare con una accuratezza elevata (almeno il 90%), i difetti eventualmente presenti su di una maglia tubolare. In questo caso è stato sviluppato un prototipo che combina un sistema di acquisizione digitale con un sistema di elaborazione dati basato su Image Processing e Reti Neurali in grado di ottemperare allo scopo prefisso. Inoltre è riportata un’esperienza analoga applicata su macchine di controllo delle difettosità (tribunali o macchine a specchio). Il sistema sviluppato ha permesso di individuare i difetti con una affidabilità del 94% ossia statisticamente maggiore della precisione richiesta in specifica. Il vantaggio che ne consegue per l’azienda tessile è di disporre di un sistema automatico e funzionante in tempo reale in grado di compiere valutazioni “intelligenti” sullo stato della maglia tubolare (o del tessuto) esaminato. Infine nel Capitolo 6 è riportato il terzo caso di studio il cui obiettivo è quello di individuare una metodologia, senza contatto ed operante in tempo reale, che consenta la misura di alcuni parametri relativi ai tessuti sottoposti alla garzatura. In particolare si sviluppa una tecnica in grado di misurare lungo la linea grandezze che, fino ad ora, sono estratte solo sperimentalmente all’interno di laboratori specializzati. Le misure, validate da tecniche sperimentali, si sono rivelate estremamente accurate ed attendibili.