J. Anitha

A pattern based PSO approach for block matching in motion estimation

Block matching motion estimation is a popular method in developing video coding applications. A new algorithm has been proposed for reducing the number of search points using a pattern based particle swarm optimization (PSO) for motion estimation. The conventional particle swarm optimization has been modified to provide accurate solutions in motion estimation problems. This leads to very low computational cost and good estimation accuracy. Due to the center biased nature of the videos, the proposed approach uses an initial pattern to speed up the convergence of the algorithm. Simulation results show that improvements over other fast block matching motion estimation algorithms could be achieved with 31%~63% of search point reduction, without degradation of image quality.

Enhanced modified orthogonal search for motion estimation

Motion estimation is an important part of video encoding systems. This paper presents a novel enhanced modified orthogonal search (EMOS) algorithm for block based motion estimation. The performance of this algorithm is evaluated with various video sequences and the results are compared to a traditional well-known full search algorithm (FSA), three step search (3SS), orthogonal search (OS), and modified orthogonal search (MOS). This paper introduces a half way stop technique to reduce computational complexity in the existing OS & MOS. The centre biased search pattern is facilitated for small motions. The enhanced modified orthogonal based search pattern is investigated in comparison with orthogonal and modified orthogonal search pattern and demonstrates significant speedup gain over them. The proposed EMOS algorithm can find the same motion vector with fewer search points than the OS & MOS algorithm. The quality of the reconstructed frame is comparable with that of the full search method, 3SS, OS & MOS. The advantage of proposed motion estimation technique is further justified by experimental results.

A wavelet based morphological mass detection and classification in mammograms

This paper presents an efficient mass detection and classification in mammogram images with the use of features extracted from the mass regions obtained by the automatic morphological based segmentation method. In this approach, the mammogram images are preprocessed to extract the breast profile and improve the contrast. The segmentation is done with combination of various morphological operations. In this approach, the wavelet features are extracted from the detected mass regions and is compared with feature extracted using Gray Level Co-occurrence Matrix (GLCM) to differentiate the TP and FP regions. Classifications of the mass regions are carried out through the Support Vector Machine (SVM) to separate the segmented regions into masses and non-masses based on the features. The methodology achieves 95% of accuracy.

A dual stage adaptive thresholding (DuSAT) for automatic mass detection in mammograms

BACKGROUND AND OBJECTIVE Early detection and diagnosis of breast cancer through mammography screening reduces breast cancer mortality by around 20%. However it is often a complex process to differentiate abnormalities due to the ill-defined margins and subtle appearances. METHOD This paper investigates a new computer aided approach to detect the abnormalities in the digital mammograms using a Dual Stage Adaptive Thresholding (DuSAT). The suspicious mass region is identified using global histogram and local window thresholding method. The global thresholding is done based on the Histogram Peak Analysis (HPA) of the entire image and the threshold is obtained by maximizing the proposed threshold selection criteria. The local thresholding is carried out for each pixel in a defined neighborhood window that provides precise segmentation results. RESULTS The algorithm is verified with 300 images in the DDSM database and 170 images in the mini-MIAS database. Experimental results show that the proposed algorithm achieves an average sensitivity of 92.5% with 1.06 FP/image for DDSM database and an average sensitivity of 93.5% with 0.62 FP/image for mini-MIAS database. CONCLUSION The achieved results depict that the proposed approach provides better results compared to other state-of-art methods for mass detection that helps the radiologists in diagnosis of breast cancer at early stage.

A hierarchical algorithm with fast convergence spiral search pattern for block matching in motion estimation

In a video encoding process, motion estimation ME has an important effect, because it approximately takes 60% to 80% of computation in the total computation complexity of video coding. ME is usually applied to reduce the redundancy that exists between successive frames of a video sequence. The motion vector MV obtained through this process represent the movement of objects between the frames. One of the most efficient and well-known ME techniques is block matching algorithm BMA. BMAs have been widely used in several video coding standards. There are various BMAs which try to reduce the number of computations in order to speed up the video encoding process. Here, a block matching ME algorithm called hierarchical algorithm with fast convergence spiral search HAFCSS is developed with the intention of reducing the computational complexity while maintaining the visual quality. This is implemented in MATLAB with different video sequences having different motion content. The proposed method provides a better speed improvement of 9% to 45% with reduced number of search points.

A spatial fuzzy based level set method for mammogram mass segmentation

The evolution of level set segmentation needs an appropriate initialization and controlling parameters, which requires manual intervention. A new spatial fuzzy level set algorithm is proposed in this paper to facilitate the automatic mammogram image segmentation. The initial segmentation of the mass is obtained by spatial fuzzy clustering that incorporates the spatial information and local intensity information to compute the weighted summed image. The maximum membership cluster extracted from the fuzzy clustering is used as an initial contour for the level set segmentation that is used to refine the mass boundary. The control parameters of level set algorithm also estimated from the results of fuzzy clustering. Experimental results show that the proposed method gives promising results to effectively segment the mass regions in the mammograms.