Yee-Hong Yang

Multiresolution color image segmentation

Image segmentation is the process by which an original image is partitioned into some homogeneous regions. In this paper, a novel multiresolution color image segmentation (MCIS) algorithm which uses Markov random fields (MRFs) is proposed. The proposed approach is a relaxation process that converges to the MAP (maximum a posteriori) estimate of the segmentation. The quadtree structure is used to implement the multiresolution framework, and the simulated annealing technique is employed to control the splitting and merging of nodes so as to minimize an energy function and therefore, maximize the MAP estimate. The multiresolution scheme enables the use of different dissimilarity measures at different resolution levels. Consequently, the proposed algorithm is noise resistant. Since the global clustering information of the image is required in the proposed approach, the scale space filter (SSF) is employed as the first step. The multiresolution approach is used to refine the segmentation. Experimental results of both the synthesized and real images are very encouraging. In order to evaluate experimental results of both synthesized images and real images quantitatively, a new evaluation criterion is proposed and developed. >

First Sight: A human body outline labeling system

First Sight, a vision system in labeling the outline of a moving human body, is proposed in this paper. The emphasis of First Sight is on the analysis of motion information gathered solely from the outline of a moving human object. Two main processes are implemented in First Sight. The first process uses a novel technique to extract the outline of a moving human body from an image sequence. The second process, which employs a new human body model, interprets the outline and produces a labeled two-dimensional human body stick figure for each frame of the image sequence. Extensive knowledge of the structure, shape, and posture of the human body is used in the model. The experimental results of applying the technique on unedited image sequences with self-occlusions and missing boundary lines are encouraging. >

The background primal sketch: an approach for tracking moving objects

In this paper we present an algorithm that integrates spatial and temporal information for the tracking of moving nonrigid objects. In addition, we obtain outlines of the moving objects.Three basic ingredients are employed in the proposed algorithm, namely, the background primal sketch, the threshold, and outlier maps. The background primal sketch is an edge map of the background without moving objects. If the background primal sketch is known, then edges of moving objects can be determined by comparing the edge map of the input image with the background primal sketch. A moving edge point is modeled as an outlier, that is, a pixel with an edge value differing from the background edge value in the background primal sketch by an amount larger than the threshold in the threshold map at the same physical location. The map that contains all the outliers is called the outlier map. In this paper we present techniques based on robust statistics for determining the background primal sketch, the threshold, and outlier maps.In an ideal situation the outlier map would contain the complete outlines of the moving objects. In practice, the outliers do not form closed contours. The final step of the algorithm employs an edge-guided morphological approach to generate closed outlines of the moving objects. The proposed approach has been tested on sequences of moving human blood cells (neutrophil) as well as of human body motion with encouraging results.

Near real-time reliable stereo matching using programmable graphics hardware

A near-real-time stereo matching technique is presented in this paper, which is based on the reliability-based dynamic programming algorithm we proposed earlier. The new algorithm can generate semi-dense disparity maps using only two dynamic programming passes, while our previous approach requires 20-30 passes. We also implement the algorithm on programmable graphics hardware, which further improves the processing speed. The experiments on the four Middlebury stereo datasets show that the new algorithm can produce dense (>85% of the pixels) and reliable (error rate <0.3%) matches in near real-time (0.05-0.1 sec). If needed, it can also be used to generate dense disparity maps. Based on the evaluation conducted by the Middlebury Stereo Vision Research Website, the new algorithm is ranked between the variable window and the graph cuts approaches and currently is the most accurate dynamic programming based technique. When more than one reference images are available, the accuracy can be further improved with little extra computation time.

Face recognition approach based on rank correlation of Gabor-filtered images

Face recognition is challenging because variations can be introduced to the pattern of a face by varying pose, lighting, scale, and expression. A new face recognition approach using rank correlation of Gabor-filtered images is presented. Using this technique, Gabor filters of different sizes and orientations are applied on images before using rank correlation for matching the face representation. The representation used for each face is computed from the Gabor-filtered images and the original image. Although training requires a fairly substantial length of time, the computation time required for recognition is very short. Recognition rates ranging between 83.5% and 96% are obtained using the AT&T (formerly ORL) database using different permutations of 5 and 9 training images per subject. In addition, the effect of pose variation on the recognition system is systematically determined using images from the UMIST database.

Region-Tree Based Stereo Using Dynamic Programming Optimization

In this paper, we present a novel stereo algorithm that combines the strengths of region-based stereo and dynamic programming on a tree approaches. Instead of formulating an image as individual scan-lines or as a pixel tree, a new region tree structure, which is built as a minimum spanning tree on the adjacency-graph of an over-segmented image, is used for the global dynamic programming optimization. The resulting disparity maps do not contain any streaking problem as is common in scanline-based algorithms because of the tree structure. The performance evaluation using the Middlebury benchmark datasets shows that the performance of our algorithm is comparable in accuracy and efficiency with top ranking algorithms.

Stationary background generation: an alternative to the difference of two images

Abstract Image understanding is the process of determining the contents of an image. In many cases, only specific portions of the image are of interest. Motion is one way to define the portions of the image that are of interest: anything that moves should be identified. One commonly used approach is to employ the difference of two consecutive images. In this paper, a complementary approach which is to determine the stationary background first is proposed. Once the background is known, the moving objects can be easily located.

Theoretical analysis of illumination in PCA-based vision systems

In this paper, we address the problem: How to account for arbitrary illumination effects for a pose of an object in PCA-based vision systems. This is a key problem since after solving this problem, the approach can be applied directly to an arbitrary number of poses of an arbitrary number of objects. We solve this problem by first generating an analytic closed-form formula of the covariance matrix for a special lighting condition. Then after analyzing all possible illumination effects, an equation called the illumination equation is derived to account for arbitrary illumination effects. Experiments on simulated conditions and real world conditions confirm the advantages of our new methods. A direct application of current research is that for any pose, our method can be used to compress the image of the object in any possible illumination. This is demonstrated in the real world experiment in the paper. Furthermore, this paper gives a new framework on how to address illumination effects in computer vision in general.

Multiresolution texture segmentation with application to diagnostic ultrasound images

A multiresolution texture segmentation (MTS) approach to image segmentation that addresses the issues of texture characterization, image resolution, and time to complete the segmentation is presented. The approach generalizes the conventional simulated annealing method to a multiresolution framework and minimizes an energy function that is dependent on the resolution of the size of the texture blocks in an image. A rigorous experimental procedure is also proposed to demonstrate the advantages of the proposed MTS approach on the accuracy of the segmentation, the efficiency of the algorithm, and the use of varying features at different resolution. Semireal images, created by sampling a series of diagnostic ultrasound images of an ovary in vitro, were tested to produce statistical measures on the performance of the approach. The ultrasound images themselves were then segmented to determine if the approach can achieve accurate results for the intended ultrasound application. Experimental results suggest that the MTS approach converges faster and produces better segmentation results than the single-level approach.

Dynamic two-strip algorithm in curve fitting

Abstract In this paper, a new dynamic strip algorithm for fitting a curve with lines is presented. The algorithm first finds the best fitted left-hand side and right-hand side strips at each point on the curve. A figure of merit is computed based on the fitting results. Then, a local maximum detection process is applied to pick out points of high curvature. The approximated curve is one with the high curvature points connected by straight lines. The global structure of a shape, which is insensitive to a scale and orientation changes, is obtained by merging the results from several runs of the algorithm with different minimum strip widths.