Learning Grasp Affordances from Vision

Monimutkaisissa ja muuttuvissa ympäristöissä työskentelevät robotit tarvitsevat kykyä manipuloida ja tarttua esineisiin. Tämä työ tutkii robottitarttumisen ja robottitartuntapis-teiden koneoppimisen aiempaa tutkimusta ja nykytilaa. Nykyaikaiset menetelmät käydään läpi, ja Le:n koneoppimiseen pohjautuva luokitin toteutetaan, koska se tarjoaa parhaan onnistumisprosentin tutkituista menetelmistä ja on muokattavissa sopivaksi käytettävissä olevalle robotille. Toteutettu menetelmä käyttää intensititeettikuvaan ja syvyyskuvaan po-hjautuvia ominaisuuksi luokitellakseen potentiaaliset tartuntapisteet. Tämän toteutuksen tulokset esitellään.

Neural mechanisms underlying conscious and unconscious vision. Evidence from event-related potentials and transcranial magnetic stimulation

Vision affords us with the ability to consciously see, and use this information in our behavior. While research has produced a detailed account of the function of the visual system, the neural processes that underlie conscious vision are still debated. One of the aims of the present thesis was to examine the time-course of the neuroelectrical processes that correlate with conscious vision. The second aim was to study the neural basis of unconscious vision, that is, situations where a stimulus that is not consciously perceived nevertheless influences behavior. According to current prevalent models of conscious vision, the activation of visual cortical areas is not, as such, sufficient for consciousness to emerge, although it might be sufficient for unconscious vision. Conscious vision is assumed to require reciprocal communication between cortical areas, but views differ substantially on the extent of this recurrent communication. Visual consciousness has been proposed to emerge from recurrent neural interactions within the visual system, while other models claim that more widespread cortical activation is needed for consciousness. Studies I-III compared models of conscious vision by studying event-related potentials (ERP). ERPs represent the brain’s average electrical response to stimulation. The results support the model that associates conscious vision with activity localized in the ventral visual cortex. The timing of this activity corresponds to an intermediate stage in visual processing. Earlier stages of visual processing may influence what becomes conscious, although these processes do not directly enable visual consciousness. Late processing stages, when more widespread cortical areas are activated, reflect the access to and manipulation of contents of consciousness. Studies IV and V concentrated on unconscious vision. By using transcranial magnetic stimulation (TMS) we show that when early visual cortical processing is disturbed so that subjects fail to consciously perceive visual stimuli, they may nevertheless guess (above chance-level) the location where the visual stimuli were presented. However, the results also suggest that in a similar situation, early visual cortex is necessary for both conscious and unconscious perception of chromatic information (i.e. color). Chromatic information that remains unconscious may influence behavioral responses when activity in visual cortex is not disturbed by TMS. Our results support the view that early stimulus-driven (feedforward) activation may be sufficient for unconscious processing. In conclusion, the results of this thesis support the view that conscious vision is enabled by a series of processing stages. The processes that most closely correlate with conscious vision take place in the ventral visual cortex ~200 ms after stimulus presentation, although preceding time-periods and contributions from other cortical areas such as the parietal cortex are also indispensable. Unconscious vision relies on intact early visual activation, although the location of visual stimulus may be unconsciously resolved even when activity in the early visual cortex is interfered with.

From Persephone to Pan: D. H. Lawrence's mythopoeic vision of the integrated personality with special emphasis on the short fiction and other writings in the early nineteen twenties

This doctoral thesis was published in printed form in 1987. It was digitized from paper copy in 2013. Unfortunately on some pages the digitizaion process has not been complete, i.e there are some minor typographic erros on some pages.

Experimental validation of a 3-D vision-based measurement system applied to robot calibration

One of the problems that slows the development of off-line programming is the low static and dynamic positioning accuracy of robots. Robot calibration improves the positioning accuracy and can also be used as a diagnostic tool in robot production and maintenance. A large number of robot measurement systems are now available commercially. Yet, there is a dearth of systems that are portable, accurate and low cost. In this work a measurement system that can fill this gap in local calibration is presented. The measurement system consists of a single CCD camera mounted on the robot tool flange with a wide angle lens, and uses space resection models to measure the end-effector pose relative to a world coordinate system, considering radial distortions. Scale factors and image center are obtained with innovative techniques, making use of a multiview approach. The target plate consists of a grid of white dots impressed on a black photographic paper, and mounted on the sides of a 90-degree angle plate. Results show that the achieved average accuracy varies from 0.2mm to 0.4mm, at distances from the target from 600mm to 1000mm respectively, with different camera orientations.

Performance evaluation of 3D computer vision techniques

This work presents the implementation and comparison of three different techniques of three-dimensional computer vision as follows: • Stereo vision - correlation between two 2D images • Sensorial fusion - use of different sensors: camera 2D + ultrasound sensor (1D); • Structured light The computer vision techniques herein presented took into consideration the following characteristics: • Computational effort ( elapsed time for obtain the 3D information); • Influence of environmental conditions (noise due to a non uniform lighting, overlighting and shades); • The cost of the infrastructure for each technique; • Analysis of uncertainties, precision and accuracy. The option of using the Matlab software, version 5.1, for algorithm implementation of the three techniques was due to the simplicity of their commands, programming and debugging. Besides, this software is well known and used by the academic community, allowing the results of this work to be obtained and verified. Examples of three-dimensional vision applied to robotic assembling tasks ("pick-and-place") are presented.

Machine vision methods for process measurements in pulping

The papermaking industry has been continuously developing intelligent solutions to characterize the raw materials it uses, to control the manufacturing process in a robust way, and to guarantee the desired quality of the end product. Based on the much improved imaging techniques and image-based analysis methods, it has become possible to look inside the manufacturing pipeline and propose more effective alternatives to human expertise. This study is focused on the development of image analyses methods for the pulping process of papermaking. Pulping starts with wood disintegration and forming the fiber suspension that is subsequently bleached, mixed with additives and chemicals, and finally dried and shipped to the papermaking mills. At each stage of the process it is important to analyze the properties of the raw material to guarantee the product quality. In order to evaluate properties of fibers, the main component of the pulp suspension, a framework for fiber characterization based on microscopic images is proposed in this thesis as the first contribution. The framework allows computation of fiber length and curl index correlating well with the ground truth values. The bubble detection method, the second contribution, was developed in order to estimate the gas volume at the delignification stage of the pulping process based on high-resolution in-line imaging. The gas volume was estimated accurately and the solution enabled just-in-time process termination whereas the accurate estimation of bubble size categories still remained challenging. As the third contribution of the study, optical flow computation was studied and the methods were successfully applied to pulp flow velocity estimation based on double-exposed images. Finally, a framework for classifying dirt particles in dried pulp sheets, including the semisynthetic ground truth generation, feature selection, and performance comparison of the state-of-the-art classification techniques, was proposed as the fourth contribution. The framework was successfully tested on the semisynthetic and real-world pulp sheet images. These four contributions assist in developing an integrated factory-level vision-based process control.

Weed mapping using a machine vision system

Weed mapping is a useful tool for site-specific herbicide applications. The objectives of this study were (1) to determine the percentage of land area covered by weeds in no-till and conventionally tilled fields of common bean using digital image processing and geostatistics, and (2) to compare two types of cameras. Two digital cameras (color and infrared) and a differential GPS were affixed to a center pivot structure for image acquisition. Sample field images were acquired in a regular grid pattern, and the images were processed to estimate the percentage of weed cover. After calculating the georeferenced weed percentage values, maps were constructed using geostatistical techniques. Based on the results, color images are recommended for mapping the percentage of weed cover in no-till systems, while infrared images are recommended for weed mapping in conventional tillage systems.

Impaired regeneration of dystrophin-deficient muscle fibers is caused by exhaustion of myogenic cells

Duchenne muscular dystrophy is one of the most devastating myopathies. Muscle fibers undergo necrosis and lose their ability to regenerate, and this may be related to increased interstitial fibrosis or the exhaustion of satellite cells. In this study, we used mdx mice, an animal model of Duchenne muscular dystrophy, to assess whether muscle fibers lose their ability to regenerate after repeated cycles of degeneration-regeneration and to establish the role of interstitial fibrosis or exhaustion of satellite cells in this process. Repeated degenerative-regenerative cycles were induced by the injection of bupivacaine (33 mg/kg), a myotoxic agent. Bupivacaine was injected weekly into the right tibialis anterior muscle of male, 8-week-old mdx (N = 20) and C57Bl/10 (control, N = 10) mice for 20 and 50 weeks. Three weeks after the last injection, the mice were killed and the proportion of regenerated fibers was counted and reported as a fibrosis index. Twenty weekly bupivacaine injections did not change the ability of mdx muscle to regenerate. However, after 50 weekly bupivacaine injections, there was a significant decrease in the regenerative response. There was no correlation between the inability to regenerate and the increase in interstitial fibrosis. These results show that after prolonged repeated cycles of degeneration-regeneration, mdx muscle loses its ability to regenerate because of the exhaustion of satellite cells, rather than because of an increase in interstitial fibrosis. This finding may be relevant to cell and gene therapy in the treatment of Duchenne muscular dystrophy.

Vision, image, record : a cultivation of the visual field

I den första delen av den här avhandlingen presenteras en bildens genealogi. Den skildrar hur begreppen för bilden, seendet och jaget utvecklades i relation till varandra i en specifik vetenskaplig och filosofisk kontext. Berättelsen sträcker sig från den tidiga renässansen och det perspektivistiska måleriet, till fotografiets födelse och positivismen. Den här utvecklingen medförde en form av reduktionism i vilken jagets roll – betydelsen av den mänskliga psykologin, vårt omdöme, vår uppmärksamhet och vår vilja – blev förbisedd. Inom den här tanketraditionen uppstod en förskjutning, från en förståelse av bilden som en representation av det tredimensionella rummet på en tvådimensionell yta, till en uppfattning om bilden som en genomskinlig ruta, ett fönster ut mot världen. Idén om avbildningen som en neutral ”blick från ingenstans” kom att förstärka en skeptisk hållning till kommunikation, dialog och vittnesmål och därmed även undergräva vår tillit till varandra och följaktligen vår tillit till oss själva. I den andra delen erbjuder författaren ett alternativ till den tanketradition som behandlas i den första delen. Det som blev förbisett i uppfattningen om en blick från ingenstans var att bilden är ett hjälpmedel då vi bearbetar vårt synfält. Bilden hjälper oss att dela vår syn på saker. Genom den här uppgiften av att dela blir bilden riktningsgivande i våra försök att orientera oss i världen. Jag kan stå bredvid en annan människa och se vad hon ser, men jag vet inte nödvändigtvis hur hon uppfattar det vi ser. Bilden lägger till ett led i det här förhållandet eftersom den inte enbart visar vad den andra ser. När bilden fungerar som den skall visar den också hur den andra ser och på det här sättet blir bilden verksam. Den föreliggande avhandlingen kombinerar epistemologi med vetenskapshistoria och visuella kulturstudier, men dess huvudintresse är filosofiskt. Den befattar sig med filosofiska missförstånd angående avbildning som en mimetisk konstform, kunskap som domesticering och varseblivning som mottagning av data. ------------------------------------------------------ Tämän väitöskirjan ensimmäinen osa selvittää kuvakäsitteen genealogiaa. Se havainnollistaa miten kuvan, näkemisen ja minän käsitteet kehittyivät suhteessa toisiinsa. Kertomus ulottuu varhaisesta renessanssista ja perspektivistisestä maalaustaiteesta, positivismin aikakauteen ja valokuvan syntyyn. Tämä kehitys toi mukanaan reduktionismin jossa minän rooli – ihmisen psykologian merkitys, meidän arviointikyky, meidän huomiokyky sekä meidän tahtomme – vaipui unohduksiin. Ajatusmaailmassa tapahtui siirtymä, kuvan merkitys vaihtui käsityksestä jossa se on kolmiulotteisen tilan representaatio kaksiulotteisella pinnalla, käsitykseen jossa kuva on läpinäkyvä ruutu, ikkuna kohti maailmaa. Ajatus kuvasta neutraalin näkökulman kantajana vahvisti skeptistä suhtautumista kommunikaatiota, dialogisuutta ja subjektiivisuutta kohtaan. Tämä skeptisyys ilmentyi myös vahvana epäluottamuksena ihmiskeskeisyyttä ja toiseutta kohtaan. Toisessa osassa tekijä tarjoaa vaihtoehdon tälle skeptiselle ajatusmaailmalle jota tarkastellaan ensimmäisessä osassa. Kuva on myös väline joka auttaa meitä jäsentämään meidän näkökenttäämme. Se auttaa meitä jakamaan meidän käsityksiä toistemme kanssa. Tämä näkemisen jakamisen käytäntö on kuvan keskeinen tehtävä. Voin seistä toisen ihmisen vieressä ja nähdä samat asiat kuin hän, mutta en välttämättä ymmärrä miten hän näkee nämä asiat. Kuva lisää jotain olennaista tähän suhteeseen. Kun kuva toimii niin kun sen kuuluu toimia, se näyttää myös miten toinen näkee, tällä tavalla kuvasta tulee välittäjä. Tämä väitöskirja yhdistää epistemologiaa, tieteen historiaa ja visuaalisen kulttuurin tutkimusta, mutta sen pääasiallinen tavoite on filosofinen. Se käsittelee filosofisia väärinkäsityksiä koskien kuvan eideettisyyttä.

Impaired beta-adrenergic response and decreased L-type calcium current of hypertrophied left ventricular myocytes in postinfarction heart failure

Infarct-induced heart failure is usually associated with cardiac hypertrophy and decreased ß-adrenergic responsiveness. However, conflicting results have been reported concerning the density of L-type calcium current (I Ca(L)), and the mechanisms underlying the decreased ß-adrenergic inotropic response. We determined I Ca(L) density, cytoplasmic calcium ([Ca2+]i) transients, and the effects of ß-adrenergic stimulation (isoproterenol) in a model of postinfarction heart failure in rats. Left ventricular myocytes were obtained by enzymatic digestion 8-10 weeks after infarction. Electrophysiological recordings were obtained using the patch-clamp technique. [Ca2+]i transients were investigated via fura-2 fluorescence. ß-Adrenergic receptor density was determined by [³H]-dihydroalprenolol binding to left ventricle homogenates. Postinfarction myocytes showed a significant 25% reduction in mean I Ca(L) density (5.7 ± 0.28 vs 7.6 ± 0.32 pA/pF) and a 19% reduction in mean peak [Ca2+]i transients (0.13 ± 0.007 vs 0.16 ± 0.009) compared to sham myocytes. The isoproterenol-stimulated increase in I Ca(L) was significantly smaller in postinfarction myocytes (Emax: 63.6 ± 4.3 vs 123.3 ± 0.9% in sham myocytes), but EC50 was not altered. The isoproterenol-stimulated peak amplitude of [Ca2+]i transients was also blunted in postinfarction myocytes. Adenylate cyclase activation through forskolin produced similar I Ca(L) increases in both groups. ß-Adrenergic receptor density was significantly reduced in homogenates from infarcted hearts (Bmax: 93.89 ± 20.22 vs 271.5 ± 31.43 fmol/mg protein in sham myocytes), while Kd values were similar. We conclude that postinfarction myocytes from large infarcts display reduced I Ca(L) density and peak [Ca2+]i transients. The response to ß-adrenergic stimulation was also reduced and was probably related to ß-adrenergic receptor down-regulation and not to changes in adenylate cyclase activity.

Hypercaloric cafeteria-like diet induced UCP3 gene expression in skeletal muscle is impaired by hypothyroidism

The uncoupling protein UCP3 belongs to a family of mitochondrial carriers located in the inner mitochondrial membrane of certain cell types. It is expressed almost exclusively at high levels in skeletal muscle and its physiological role has not been fully determined in this tissue. In the present study we have addressed the possible interaction between a hypercaloric diet and thyroid hormone (T3), which are strong stimulators of UCP3 gene expression in skeletal muscle. Male Wistar rats weighing 180 ± 20 g were rendered hypothyroid by thyroidectomy and the addition of methimazole (0.05%; w/v) to drinking water after surgery. The rats were fed a hypercaloric cafeteria diet (68% carbohydrates, 13% protein and 18% lipids) for 10 days and sacrificed by decapitation. Subsequently, the gastrocnemius muscle was dissected, total RNA was isolated with Trizol™ and UCP3 gene expression was determined by Northern blotting using a specific probe. Statistical analysis was performed by one-way analysis of variance (ANOVA) followed by the Student-Newman-Keuls post-test. Skeletal muscle UCP3 gene expression was decreased by 60% in hypothyroid rats and UCP3 mRNA expression was increased 70% in euthyroid cafeteria-fed rats compared to euthyroid chow-fed animals, confirming previous studies. Interestingly, the cafeteria diet was unable to stimulate UCP3 gene expression in hypothyroid animals (40% lower as compared to euthyroid cafeteria-fed animals). The results show that a hypercaloric diet is a strong stimulator of UCP3 gene expression in skeletal muscle and requires T3 for an adequate action.

Prevalence of the A1555G (12S rRNA) and tRNA Ser(UCN) mitochondrial mutations in hearing-impaired Brazilian patients

Mitochondrial mutations are responsible for at least 1% of the cases of hereditary deafness, but the contribution of each mutation has not yet been defined in African-derived or native American genetic backgrounds. A total of 203 unselected hearing-impaired patients were screened for the presence of the mitochondrial mutation A1555G in the 12S rRNA gene and mutations in the tRNA Ser(UCN) gene in order to assess their frequency in the ethnically admixed Brazilian population. We found four individuals with A1555G mutation (2%), which is a frequency similar to those reported for European-derived populations in unselected samples. On the other hand, complete sequencing of the tRNA Ser(UCN) did not reveal reported pathogenic substitutions, namely A7445G, 7472insC, T7510C, or T7511C. Instead, other rare substitutions were found such as T1291C, A7569G, and G7444A. To evaluate the significance of these findings, 110 "European-Brazilians" and 190 "African-Brazilians" unrelated hearing controls were screened. The T1291C, A7569G and G7444A substitutions were each found in about 1% (2/190) of individuals of African ancestry, suggesting that they are probably polymorphic. Our results indicate that screening for the A1555G mutation is recommended among all Brazilian deaf patients, while testing for mutations in the tRNA Ser(UCN) gene should be considered only when other frequent deafness-causing mutations have been excluded or in the presence of a maternal transmission pattern.

Immunoglobulin production is impaired in protein-deprived mice and can be restored by dietary protein supplementation

Most contacts with food protein and microbiota antigens occur at the level of the gut mucosa. In animal models where this natural stimulation is absent, such as germ-free and antigen-free mice, the gut-associated lymphoid tissue (GALT) and systemic immunological activities are underdeveloped. We have shown that food proteins play a critical role in the full development of the immune system. C57BL/6 mice weaned to a diet in which intact proteins are replaced by equivalent amounts of amino acids (Aa diet) have a poorly developed GALT as well as low levels of serum immunoglobulins (total Ig, IgG, and IgA, but not IgM). In the present study, we evaluated whether the introduction of a protein-containing diet in 10 adult Aa-fed C57BL/6 mice could restore their immunoglobulin levels and whether this recovery was dependent on the amount of dietary protein. After the introduction of a casein-containing diet, Aa-fed mice presented a fast recovery (after 7 days) of secretory IgA (from 0.33 to 0.75 mg/mL, while in casein-fed mice this value was 0.81 mg/mL) and serum immunoglobulin levels (from 5.39 to 10.25 mg/mL of total Ig). Five percent dietary casein was enough to promote the restoration of secretory IgA and serum immunoglobulin levels to a normal range after 30 days feeding casein diet (as in casein-fed mice - 15% by weight of diet). These data suggest that the defect detected in the immunoglobulin levels was a reversible result of the absence of food proteins as an antigenic stimulus. They also indicate that the deleterious consequences of malnutrition at an early age for some immune functions may be restored by therapeutic intervention later in life.

Long-term loss of color vision after exposure to mercury vapor

We evaluated the color vision of 24 subjects (41.6 ± 6.5 years; 6 females) who worked in fluorescent lamp industries. They had been occupationally exposed to mercury vapor (10.6 ± 5.2 years) and had been away from the source of exposure for 6.4 ± 4.04 years. Mean urinary concentration of mercury was 40.6 ± 36.4 µg/g creatinine during or up to 1 year after exposure and 2.71 ± 1.19 µg/g creatinine at the time of color vision testing or up to 1 year thereafter. All patients were diagnosed with chronic mercury intoxication, characterized by clinical symptoms and neuropsychological alterations. A control group (N = 36, 48.6 ± 11.9 years, 10 females, 1.5 ± 0.47 µg mercury/g creatinine) was subjected to the same tests. Inclusion criteria for both groups were Snellen VA 20/30 or better and absence of known ophthalmologic pathologies. Color discrimination was assessed with the Farnsworth D-15 test (D-15) and with the Lanthony D-15d test (D-15d). Significant differences were found between the two eyes of the patients (P < 0.001) in both tests. Results for the worst eye were also different from controls for both tests: P = 0.014 for D-15 and P < 0.001 for D-15d. As shown in previous studies, the D-15d proved to be more sensitive than the D-15 for the screening and diagnosis of the color discrimination losses. Since color discrimination losses were still present many years after the end of exposure, they may be considered to be irreversible, at least under the conditions of the present study.