Chung-Ming Kuo

Video object tracking using adaptive Kalman filter

In this paper, a new video moving object tracking method is proposed. In initialization, a moving object selected by the user is segmented and the dominant color is extracted from the segmented target. In tracking step, a motion model is constructed to set the system model of adaptive Kalman filter firstly. Then, the dominant color of the moving object in HSI color space will be used as feature to detect the moving object in the consecutive video frames. The detected result is fed back as the measurement of adaptive Kalman filter and the estimate parameters of adaptive Kalman filter are adjusted by occlusion ratio adaptively. The proposed method has the robust ability to track the moving object in the consecutive frames under some kinds of real-world complex situations such as the moving object disappearing totally or partially due to occlusion by other ones, fast moving object, changing lighting, changing the direction and orientation of the moving object, and changing the velocity of moving object suddenly. The proposed method is an efficient video object tracking algorithm.

A fast MPEG-7 dominant color extraction with new similarity measure for image retrieval

Dominant color descriptor (DCD) is one of the color descriptors proposed by MPEG-7 that has been extensively used for image retrieval. Among the color descriptors, DCD describes the salient color distributions in an image or a region of interest. DCD provides an effective, compact, and intuitive representation of colors presented in an image. In this paper, we will develop an efficient scheme for dominant color extraction. This approach significantly improves the efficiency of computation for dominant color extraction. In addition, we propose a modification for the MPEG-7 dissimilarity measure, which effectively improves the accuracy of perceptive similarity. Experimental results show that the proposed method achieves performance improvement not only in saving features extraction cost but also perceptually similar image retrieval.

Design of two-dimensional FIR digital filters by a two-dimensional WLS technique

For two-dimensional (2-D) FIR filter design, the conventional weighted least squares (WLS) technique rearranges the filter parameters of 2-D form into their corresponding one-dimensional (1-D) form, thus resulting in expensive computation. This paper presents a new computationally efficient WLS technique for the design of 2-D FIR filters. We introduce an updating desired frequency response which implicitly includes the weighting function such that the sum of weighted square errors to be minimized can be represented in a 2-D matrix form. This makes it possible to keep all filter parameters in their natural 2-D form, thereby reducing the computational complexity from O(N/sup G/) to O(N/sup 3/). It is confirmed through design examples that the new technique is computationally very efficient and leads to nearly optimal approximations. This technique is suitable for the design of 2-D real zero-phase FIR filters with quadrantal symmetric or antisymmetric frequency response and can also be applied to the design of 1-D FIR filters.

Motion estimation for video compression using Kalman filtering

Motion estimation plays an important role for the compression of video signals. This paper presents a new block-based motion estimation method using Kalman filtering. The new method utilizes the predicted motion and measured motion to obtain an optimal estimate of motion vector. The autoregressive models are employed to fit the motion correlation between neighboring blocks and then achieve predicted motion information. The measured motion information is obtained by the conventional block-based fast search schemes. Several algorithms based on either one- or two dimensional models using either nonadaptive or adaptive Kalman filters are developed. The analysis of computational complexity and the simulation results indicate that the proposed method achieves significant savings on computation along with smoother motion vector fields and similar picture quality, when compared to the conventional full search algorithm.

A Novel Prediction-Based Directional Asymmetric Search Algorithm for Fast Block-Matching Motion Estimation

This paper proposes a very fast block-matching motion estimation algorithm for video compression. This method uses a new concept involving a very compact center-biased characteristic in developing directional asymmetric search patterns, which we refer to as directional asymmetric search (DAS). The initial pattern of the DAS is a compact cross pattern containing only five initial search points. The DAS utilizes error information (block distortions) of the search patterns to determine the search direction, and then asymmetric search patterns are used in the subsequent steps accordingly. Furthermore, a prediction scheme and a best match prejudgment scheme are incorporated to favor fast motion and to benefit stationary and quasi-stationary blocks, respectively. Therefore, the proposed method significantly reduces the number of search points for locating a motion vector. Compared to conventional fast algorithms, the proposed method has the fastest search speed and most satisfactory PSNR values for all test sequences.

Color-Based Image Salient Region Segmentation Using Novel Region Merging Strategy

In this paper, we propose a novel unsupervised algorithm for the segmentation of salient regions in color images. There are three phases in this algorithm. In the first phase, we use nonparametric density estimation to extract candidates of dominant colors in an image, which are then used for the quantization of the image. The label map of the quantized image forms initial regions of segmentation. In the second phase, we define salient region with two properties; i.e., conspicuous; compact and complete. According to the definition, two new parameters are proposed. One is called ldquoImportance indexrdquo, which is used to measure the importance of a region, and the other is called ldquoMerging likelihoodrdquo, which is utilized to measure the suitability of region merging. Initial regions are merged based on the two new parameters. In the third phase, a similarity check is performed to further merge the surviving regions. Experimental results show that the proposed method achieves excellent segmentation performance for most of our test images. In addition, the computation is very efficient.

Kalman filtering based rate-constrained motion estimation for very low bit rate video coding

The rate-constrained (R-D) motion estimation techniques have been presented to improve the conventional block-matching algorithm by using a joint rate and distortion criterion. This paper presents two motion estimation algorithms using Kalman filter to further enhance the performance of the conventional R-D motion estimation at a relative low computational cost. The Kalman filter exploits the correlation of block motion to achieve higher precision of motion estimation and compensation. In the first algorithm, the Kalman filter is utilized as a postprocessing to raise the motion compensation accuracy of the conventional R-D motion estimation. In the second algorithm, the Kalman filter is embedded into the optimization process of R-D motion estimation by defining a new R-D criterion. It further improves the rate-distortion performance significantly.

Similarity retrieval of shape images based on database classification

Abstract Shape is a key visual feature used to describe image content. This paper develops a novel shape-based similarity retrieval system based on database classification which exploits the contour and interior region of a shape efficiently. In this system, the database of shape images is categorized automatically into 11 classes by a simple contour feature. In query, the contour feature of the input image is used to decide which class the query image belongs to. Then, the possible classes are selected dynamically from the database and to form candidate sets with different priority orders. Then, ART region feature is employed to compare the query with the candidate sets according to the priority order. Instead of using the original contour of a shape image directly, we employ a rough version of the original contour for the classification of shapes. The similarity test results indicate that the proposed method improves retrieval accuracy and speed significantly, as compared to ART.

A new motion estimation algorithm for video coding using adaptive Kalman filter

This paper presents a new motion estimation algorithm to improve the performance of the existing searching algorithms at a relatively low computational cost. We try to amend the incorrect and/or inaccurate estimate of motion with higher precision by using Kalman filter. We first obtain a measurement of motion vector of a block by using the existing searching scheme. We then generate the predicted motion vector utilizing the inter-block correlation in both spatial and temporal directions. In general, the motion correlation in spatial direction is different from that in temporal direction. Therefore, if we appropriately utilize the motion correlation in spatial direction and temporal direction, the better performance would be achieved. In this paper, we will propose an adaptive Kalman filter which utilizes a model switching mechanism to select correct motion model. Simulation results show that the proposed technique can efficiently improve the motion estimation performance.

An efficient motion estimation algorithm for video coding using Kalman filter

This paper presents a new motion estimation algorithm to improve the performance of the existing searching algorithms at a relatively low computational cost. We try to amend the incorrect and/or inaccurate estimate of motion with higher precision by using Kalman filter. We first obtain a measurement of motion vector of a block by using the existing searching scheme. We then generate the predicted motion vector utilizing the inter-block correlation in both spatial and temporal directions. Based on these two motion information, a three-dimensional (3D) motion model is developed and then a 3D local Kalman filter is designed to obtain the optimal estimate of motion vector.