A contribution to semantic description of images and videos: an application of soft biometrics

Humans have a high ability to extract visual data information acquired by sight. Trought a learning process, which starts at birth and continues throughout life, image interpretation becomes almost instinctively. At a glance, one can easily describe a scene with reasonable precision, naming its main components. Usually, this is done by extracting low-level features such as edges, shapes and textures, and associanting them to high level meanings. In this way, a semantic description of the scene is done. An example of this, is the human capacity to recognize and describe other people physical and behavioral characteristics, or biometrics. Soft-biometrics also represents inherent characteristics of human body and behaviour, but do not allow unique person identification. Computer vision area aims to develop methods capable of performing visual interpretation with performance similar to humans. This thesis aims to propose computer vison methods which allows high level information extraction from images in the form of soft biometrics. This problem is approached in two ways, unsupervised and supervised learning methods. The first seeks to group images via an automatic feature extraction learning , using both convolution techniques, evolutionary computing and clustering. In this approach employed images contains faces and people. Second approach employs convolutional neural networks, which have the ability to operate on raw images, learning both feature extraction and classification processes. Here, images are classified according to gender and clothes, divided into upper and lower parts of human body. First approach, when tested with different image datasets obtained an accuracy of approximately 80% for faces and non-faces and 70% for people and non-person. The second tested using images and videos, obtained an accuracy of about 70% for gender, 80% to the upper clothes and 90% to lower clothes. The results of these case studies, show that proposed methods are promising, allowing the realization of automatic high level information image annotation. This opens possibilities for development of applications in diverse areas such as content-based image and video search and automatica video survaillance, reducing human effort in the task of manual annotation and monitoring.

Detecção de região saliente em imagens usando dissimilaridade de cor e amostragem por pixeis aleatórios

The purpose of this work is to demonstrate and to assess a simple algorithm for automatic estimation of the most salient region in an image, that have possible application in computer vision. The algorithm uses the connection between color dissimilarities in the image and the image’s most salient region. The algorithm also avoids using image priors. Pixel dissimilarity is an informal function of the distance of a specific pixel’s color to other pixels’ colors in an image. We examine the relation between pixel color dissimilarity and salient region detection on the MSRA1K image dataset. We propose a simple algorithm for salient region detection through random pixel color dissimilarity. We define dissimilarity by accumulating the distance between each pixel and a sample of n other random pixels, in the CIELAB color space. An important result is that random dissimilarity between each pixel and just another pixel (n = 1) is enough to create adequate saliency maps when combined with median filter, with competitive average performance if compared with other related methods in the saliency detection research field. The assessment was performed by means of precision-recall curves. This idea is inspired on the human attention mechanism that is able to choose few specific regions to focus on, a biological system that the computer vision community aims to emulate. We also review some of the history on this topic of selective attention.

Navegação de um robô móvel baseado em um modelo de consciência artificial

This work presents a study about a the Baars-Franklin architecture, which defines a model of computational consciousness, and use it in a mobile robot navigation task. The insertion of mobile robots in dynamic environments carries a high complexity in navigation tasks, in order to deal with the constant environment changes, it is essential that the robot can adapt to this dynamism. The approach utilized in this work is to make the execution of these tasks closer to how human beings react to the same conditions by means of a model of computational consci-ousness. The LIDA architecture (Learning Intelligent Distribution Agent) is a cognitive system that seeks tomodel some of the human cognitive aspects, from low-level perceptions to decision making, as well as attention mechanism and episodic memory. In the present work, a computa-tional implementation of the LIDA architecture was evaluated by means of a case study, aiming to evaluate the capabilities of a cognitive approach to navigation of a mobile robot in dynamic and unknown environments, using experiments both with virtual environments (simulation) and a real robot in a realistic environment. This study concluded that it is possible to obtain benefits by using conscious cognitive models in mobile robot navigation tasks, presenting the positive and negative aspects of this approach.

Desenvolvimento de protótipo de aplicativo móvel em Android® para o controle e acompanhamento nutricional da saúde óssea em mulheres menopáusicas

Following a drop in estrogen in the period of menopause some women begin to lose bone mass more than 1% per year reaching the end of five years with loss greater than 25%. In this regard, factors such as older age, low calcium intake and premature menopause favor the onset of osteoporosis. Preventive methods such as nutritional counseling to a proper diet and the support of technology through applications that assess dietary intake are essential. Thus, this study aimed to develop an application for Android® platform focused on the evaluation of nutritional and organic conditions involved in bone health and risks for developing osteoporosis in postmenopausal women. To achieve this goal we proceeded to a study of 72 women aged 46-79 years, from the physical exercise for bone health of the Laboratory for Research in Biochemistry and Densitometry the Federal Technological University of Paraná program. Data were collected in the second half of 2014 through tests Bone Densitometry and Body Composition, Blood Tests, Anthropometric data and Nutrition Assessment. The study included women with a current diagnosis of osteopenia or osteoporosis primary, aged more than 45 years postmenopausal. For the assessment of bone mineral density and body composition used the device Absorptiometry Dual Energy X-ray (DXA) brand Hologic Discovery TM Model A. For anthropometric assessment was included to body mass, height, abdominal circumference, Waist circumference and hip circumference. The instrument for assessing food consumption was used Recall 24 hours a day (24HR). The estimated intake of energy and nutrients was carried from the tabulation of the food eaten in the Software Diet Pro 4®. In a sub sample of 30 women with osteopenia / osteoporosis serum calcium and alkaline phosphatase tests were performed. The results demonstrated a group of women (n = 30) average calcium intake of 570mg / day (± 340). The analysis showed a mean serum calcium within the normal range (10,20mg / dl ± 0.32) and average values and slightly increased alkaline phosphatase (105.40 U / L ± 23.70). Furthermore, there was a significant correlation between the consumption of protein and the optimal daily intake of calcium (0.375 p-value 0.05). Based on these findings, we developed an application early stage in Android® platform operating system Google®, being called OsteoNutri. We chose to use Java Eclipse® where it was executed Android® version of the project; choice of application icons and setting the visual editor for building the application layouts. The DroidDraw® was used for development of the three application GUIs. For practical tests we used a cell compatible with the version that was created (4.4 or higher). The prototype was developed in conjunction with the Group and Instrumentation Applications Development (GDAI) of the Federal Technological University of Paraná. So this application can be considered an important tool in dietary control, allowing closer control consumption of calcium and dietary proteins.