Submitted by Luciana Sebin (lusebin@ufscar.br) on 2016-10-10T19:45:24Z No. of bitstreams: 1 DissKES.pdf: 1491828 bytes, checksum: c6a58b88ab03eb7890e0b182f08a3ab3 (MD5) / Approved for entry into archive by Marina Freitas (marinapf@ufscar.br) on 2016-10-13T20:15:20Z (GMT) No. of bitstreams: 1 DissKES.pdf: 1491828 bytes, checksum: c6a58b88ab03eb7890e0b182f08a3ab3 (MD5) / Approved for entry into archive by Marina Freitas (marinapf@ufscar.br) on 2016-10-13T20:15:29Z (GMT) No. of bitstreams: 1 DissKES.pdf: 1491828 bytes, checksum: c6a58b88ab03eb7890e0b182f08a3ab3 (MD5) / Made available in DSpace on 2016-10-13T20:15:42Z (GMT). No. of bitstreams: 1 DissKES.pdf: 1491828 bytes, checksum: c6a58b88ab03eb7890e0b182f08a3ab3 (MD5) Previous issue date: 2016-02-16 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Counting the number of leukocytes present in the blood vessels is an important task to understand inflammation mechanisms and to assess the effects of drugs that are being developed for the treatment of inflammatory diseases. In general, leukocyte counting is done by an observer (laboratory technician or specialist) using a sequence of intravital microscopy images obtained from blood vessel of an animal testing. However, this procedure is time-consuming, prone to errors, due to visual fatigue of the observer, and biased, due to inter and intra-observer variability. Thus, the objective of this work was the development of computational technique for the automatic detection of leukocytes in intravital video microscopy images. For this, the detection of leukocytes was performed in two main stages. In stage 1, the phase congruency measure, which is invariant to changes of contrast and lighting images, was used to calculate a blobness measure used to detect circular structures in intravital microscopy images. In stage 2, the circular structures detected in stage 1 were analyzed locally to identify only those corresponding to true leukocytes. The results were evaluated using the precision, recall and F1-measure metrics and the area under the precision-recall curves. Furthermore, the proposed technique was compared with the template matching technique. / Contar o número de leucócitos presentes em vasos sanguíneos é uma tarefa fundamental para entender mecanismos de inflamações e avaliar efeitos de drogas que estão sendo desenvolvidas para o tratamento de doenças inflamatórias. Em geral, a contagem de leucócitos é feita por um observador (técnico laboratorial ou especialista) usando uma sequência de imagens de microscopia intravital, obtidas do vaso sanguíneo de um animal de teste. Entretanto, a tarefa de contagem é demorada, propensa à erros, devido à fadiga visual do observador, e viesada, devido à variabilidade inter e intra-observador. Sendo assim, o objetivo deste trabalho foi o desenvolvimento de uma técnica computacional para a detecção automática de leucócitos em imagens de vídeo de microscopia intravital. Para isso, a detecção dos leucócitos foi realizada em dois principais estágios. No estágio 1, a medida de congruência de fase, que é invariante a mudanças de contraste e iluminação em imagens, foi utilizada para o cálculo de uma medida de blobness usada na detecção de estruturas circulares nas imagens de microscopia intravital. No estágio 2, as estruturas circulares detectadas no estágio 1 foram analisadas localmente a fim de identificar apenas aquelas correspondentes aos leucócitos verdadeiros. Os resultados foram avaliados utilizando as métricas de precisão, revocação e medida-F1 e a área sob as curvas precisão-revocação. Além disso, a técnica proposta foi comparada com a técnica casamento de padrões.

No description available.

No description available.

No description available.

Image segmentation is one of the most difficult tasks in image processing. Segmentation algorithms are generally based on searching a region where pixels share similar gray level intensity and satisfy a set of defined criteria. However, the segmented region cannot be used directly for partial image retrieval. In this dissertation, a Contour Based Image Structure (CBIS) model is introduced. In this model, images are divided into several objects defined by their bounding contours. The bounding contour structure allows individual object extraction, and partial object matching and retrieval from a standard CBIS image structure. The CBIS model allows the representation of 3D objects by their bounding contours which is suitable for parallel implementation particularly when extracting contour features and matching them for 3D images require heavy computations. This computational burden becomes worse for images with high resolution and large contour density. In this essence we designed two parallel algorithms; Contour Parallelization Algorithm (CPA) and Partial Retrieval Parallelization Algorithm (PRPA). Both algorithms have considerably improved the performance of CBIS for both contour shape matching as well as partial image retrieval. To improve the effectiveness of CBIS in segmenting images with inhomogeneous backgrounds we used the phase congruency invariant features of Fourier transform components to highlight boundaries of objects prior to extracting their contours. The contour matching process has also been improved by constructing a fuzzy contour matching system that allows unbiased matching decisions. Further improvements have been achieved through the use of a contour tailored Fourier descriptor to make translation and rotation invariance. It is proved to be suitable for general contour shape matching where translation, rotation, and scaling invariance are required. For those images which are hard to be classified by object contours such as bacterial images, we define a multi-level cosine transform to extract their texture features for image classification. The low frequency Discrete Cosine Transform coefficients and Zenike moments derived from images are trained by Support Vector Machine (SVM) to generate multiple classifiers.