Thesis (M.S.)--Rochester Institute of Technology, 2009. / Typescript. Includes bibliographical references (leaves 88-90).

Thesis (M.S.)--Youngstown State University, 2008. / Includes bibliographical references (leaves 17-19). Also available via the World Wide Web in PDF format.

Thesis (Ph.D.) - University of Queensland, 2005. / Includes bibliography.

Thesis (M.S.)--West Virginia University, 2005. / Title from document title page. Document formatted into pages; contains vii, 57 p. : ill. (some col.). Includes abstract. Includes bibliographical references (p. 53-57).

Thesis (Ph. D.)--Lehigh University, 2006. / Includes vita. Includes bibliographical references (leaves 175-189).

Localized operators, like Gabor wavelets and difference-of-Gaussian filters, are considered to be useful tools for image representation. This is due to their ability to form a ‘sparse code’ that can serve as a basis set for high-fidelity reconstruction of natural images. However, for many visual tasks, the more appropriate criterion of representational efficacy is ‘recognition’, rather than ‘reconstruction’. It is unclear whether simple local features provide the stability necessary to subserve robust recognition of complex objects. In this paper, we search the space of two-lobed differential operators for those that constitute a good representational code under recognition/discrimination criteria. We find that a novel operator, which we call the ‘dissociated dipole’ displays useful properties in this regard. We describe simple computational experiments to assess the merits of such dipoles relative to the more traditional local operators. The results suggest that non-local operators constitute a vocabulary that is stable across a range of image transformations.

Animals are sentient beings, capable of experiencing emotions. Being able to assess emotional states in farm animals is crucial to improving their welfare. Although the function of emotion is not primarily for communication, the outward expression of an emotional state involves changes in posture, vocalisations, odours and facial expressions. These changes can be perceived and used as indicators of emotional state by other animals. Since emotions can be perceived between conspecifics, understanding how emotions are identified and how they can spread within a social group could have a major impact on improving the welfare of farmed species, which are mostly reared in groups. A recently developed method for the evaluation of emotions in animals is based on cognitive biases such as judgment biases, i.e. an individual in a negative emotional state will show pessimistic judgments while and individual in a positive emotional state will show optimistic judgments. The aims of this project were to (A) establish whether sheep and goats can discriminate between images of faces of familiar conspecifics taken in different positive and negative situations, (B) establish whether sheep and goats perceive the valence (positive of negative) of the emotion expressed by the animal on the image, (C) validate the use of images of faces in cognitive bias studies. The use of images of faces of conspecifics as emotional stimuli was first validated, using a discrimination task in a two-armed maze. A new methodology was then developed across a series of experiments to assess spontaneous reactions of animals exposed to video clips or to images of faces of familiar conspecifics. Detailed observations of ear postures were used as the main behavioural indicator. Individual characteristics (dominance status within the herd, dominance pairwise relationships and humananimal relationship) were also recorded during preliminary tests and included in the analyses. The impact of a low-mood state on the perception of emotions was assessed in sheep after subjecting half of the animals to unpredictable negative housing conditions and keeping the other half in good standard housing conditions. Sheep were then presented with videos of conspecifics filmed in situations of varying valence. Reactions to ambiguous stimuli were evaluated by presenting goats with images of morphed faces. Goats were also presented with images of faces of familiar conspecifics taken situations of varying emotional intensity. Sheep could discriminate images of faces of conspecifics taken either in a negative or in a neutral situation and their learning process of the discrimination task was affected by the type of emotion displayed. Sheep reacted differently depending on the valence of the video clips (P < 0.05); however, there was no difference between the control and the low-mood groups (P > 0.05). Goats also showed different behavioural reactions to images of faces photographed in different situations (P < 0.05), indicating that they perceived the images as different. Responses to morphed images were not necessarily intermediate to responses to negative and positive images and not gradual either, which poses a major problem to the potential use of facial images in cognitive bias experiments. Overall, animals were more attentive towards images or videos of conspecifics in negative situations, i.e., presumably, in a negative emotional state. This suggests that sheep and goats are able to perceive the valence of the emotional state. The identity of the individual on the photo also affected the animals’ spontaneous reaction to the images. Social relationships such as dominance, but also affinity between the tested and photographed individual seem to influence emotion perception.

Face recognition, although being a popular area of research for over a decade has still many open research challenges. Some of these challenges include the recognition of poorly illuminated faces, recognition under pose variations and also the challenge of capturing sufficient training data to enable recognition under pose/viewpoint changes. With the appearance of cheap and effective multimodal image capture hardware, such as the Microsoft Kinect device, new possibilities of research have been uncovered. One opportunity is to explore the potential use of the depth maps generated by the Kinect as an additional data source to recognize human faces under low levels of scene illumination, and to generate new images through creating a 3D model using the depth maps and visible-spectrum/RGB images that can then be used to enhance face recognition accuracy by improving the training phase of a classification task. With the goal of enhancing face recognition, this research first investigated how depth maps, since not affected by illumination, can improve face recognition, if algorithms traditionally used in face recognition were used. To this effect a number of popular benchmark face recognition algorithms are tested. It is proved that algorithms based on LBP and Eigenfaces are able to provide high level of accuracy in face recognition due to the significantly high resolution of the depth map images generated by the latest version of the Kinect device. To complement this work a novel algorithm named the Dense Feature Detector is presented and is proven to be effective in face recognition using depth map images, in particular under wellilluminated conditions. Another technique that was presented for the goal of enhancing face recognition is to be able to reconstruct face images in different angles, through the use of the data of one frontal RGB image and the corresponding depth map captured by the Kinect, using faster and effective 3D object reconstruction technique. Using the Overfeat network based on Convolutional Neural Networks for feature extraction and a SVM for classification it is shown that a technically unlimited number of multiple views can be created from the proposed 3D model that consists features of the face if captured real at similar angles. Thus these images can be used as real training images, thus removing the need to capture many examples of a facial image from different viewpoints for the training of the image classifier. Thus the proposed 3D model will save significant amount of time and effort in capturing sufficient training data that is essential in recognition of the human face under variations of pose/viewpoint. The thesis argues that the same approach can also be used as a novel approach to face recognition, which promises significantly high levels of face recognition accuracy base on depth images. Finally following the recent trends in replacing traditional face recognition algorithms with the effective use of deep learning networks, the thesis investigates the use of four popular networks, VGG-16, VGG-19, VGG-S and GoogLeNet in depth maps based face recognition and proposes the effective use of Transfer Learning to enhance the performance of such Deep Learning networks.

No description available.

Magister Scientiae - MSc / This thesis addresses analysis, evaluation and simulation of low complexity face detection algorithms and tracking that could be used on mobile phones. Network access control using face recognition increases the user-friendliness in human-computer interaction. In order to realize a real time system implemented on handheld devices with low computing power, low complexity algorithms for face detection and face tracking are implemented. Skin color detection algorithms and face matching have low implementation complexity suitable for authentication of cellular network services. Novel approaches for reducing the complexities of these algorithms and fast implementation are introduced in this thesis. This includes a fast algorithm for face detection in video sequences, using a skin color model in the HSV (Hue-Saturation-Value) color space. It is combined with a Gaussian model of the H and S statistics and adaptive thresholds. These algorithms permit segmentation and detection of multiple faces in thumbnail images. Furthermore we evaluate and compare our results with those of a method implemented in the Chromatic Color space (YCbCr). We also test our test data on face detection method using Convolutional Neural Network architecture to study the suitability of using other approaches besides skin color as the basic feature for face detection. Finally, face tracking is done in 2D color video streams using HSV as the histogram color space. The program is used to compute 3D trajectories for a remote surveillance system. / South Africa