Browsing by Author "CHEN, LH"

Now showing 1 - 16 of 16

A 2-PHASE AREA-LEVEL LINE EDGE DETECTOR
(1992-01-01)
CHEN, LH
;
交大名義發表
;
資訊工程學系
;
National Chiao Tung University
;
Department of Computer Science
In this paper, a new feature detector is proposed to extract both edge and line points existing in a same image. It takes an n X n block as a processing unit and locates feature points through two phases: classification and extraction. In the classification phase, based on some local and global properties of the feature points, each block is classified as homogeneous, a line or an edge block. In the extraction phase, for each line or edge block, according to its block type, the proposed approach takes a corresponding feature extractor to locate all feature points. As the block type is known, the extraction work can be speeded up and the extraction result will be more accurate. Furthermore, since the type of extracted feature point is known, this will make segmentation easier. Experimental results are also given to show the effectiveness of the proposed method.
A FAST ELLIPSE/CIRCLE DETECTOR USING GEOMETRIC SYMMETRY
(1995-01-01)
HO, CT
;
CHEN, LH
;
交大名義發表
;
資訊工程學系
;
National Chiao Tung University
Through the use of a global geometric symmetry, a fast ellipse/circle detector is proposed in this paper. Based on the geometric symmetry, the proposed method first locates candidates of ellipse and circle centers. In the meantime, according to these candidate centers, all feature points in an input image are grouped into several subimages. Then, for each subimage, by using geometric symmetry again, all ellipses and circles are obtained. The method significantly reduces the time required to evaluate all possible parameters without using edge direction information. Experimental results are given to show the correctness and effectiveness of the proposed method.
A FAST SEARCH ALGORITHM FOR VECTOR QUANTIZATION USING MEAN PYRAMIDS OF CODEWORDS
(1995-02-01)
LEE, CH
;
CHEN, LH
;
交大名義發表
;
資訊工程學系
;
National Chiao Tung University
One of the most serious problems for vector quantization, especially for high dimensional vectors, is the high computational complexity of searching for the closest codeword in the codebook design and encoding phases. Although quantizing high dimensional vectors rather than low dimensional vectors results in better performance, the computation time needed for vector quantization grows exponentially with the vector dimension. This makes high dimensional vectors unsuitable for vector quantization. To overcome this problem, a fast search algorithm, under the assumption that the distortion is measured by the squared Euclidean distance, will be proposed. Using the mean pyramids of codewords, the algorithm can reject many codewords that are impossible matches and hence save a great deal of computation time. The algorithm is efficient for high dimensional codeword searches. Experimental results confirm the effectiveness of the proposed method.
A NEW APPROACH FOR RECOGNITION MULTIFONT CHINESE CHARACTERS USED IN A SPECIAL APPLICATION
(1992)
CHEN, LH
;
LEE, YK
;
資訊工程學系
;
Department of Computer Science
A NEW IMAGE COMPRESSION METHOD USING DYNAMIC PREDICTOR BASED ON CURRENT CONTEXT
(1991-07-01)
CHEN, LH
;
WENG, SF
;
交大名義發表
;
資訊工程學系
;
National Chiao Tung University
In this paper, a new image compression method is proposed. The approach first provides a decomposition method to segment the gray values of an image into several parts called buckets. The decomposition rule is to make, as much as possible, each bucket contain only one nearly-homogeneous region or edge area. For each pixel P, if the gray values of its three neighbors (top-left, top, left) belong to buckets B1, B2, and B3, respectively, then (B1, B2, B3) is referred to as P's context. An individual predictor is then derived for each set of pixels with the same context. The gray value of each pixel is then estimated by the predictor corresponding to its context. Some degree of distortion is allowed. A pixel with prediction error less than a preset value is considered as a redundant point. An image is finally coded by the prediction errors of nonredundant pixels and by the number of consecutive redundant pixels. Both of these two kinds of values are quantized and coded by their own adaptive Huffman codes. Experimental results show that the proposed method not only possesses high-speed computing property but also provides high compression ratio and good quality.
A NEW METHOD FOR CIRCULAR OBJECT DETECTION AND LOCATION
(1990-10-01)
CHEN, LH
;
LEE, KL
;
交大名義發表
;
資訊工程學系
;
National Chiao Tung University
A NEW NONITERATIVE APPROACH FOR CLUSTERING
(1994-02-01)
YIN, PY
;
CHEN, LH
;
資訊工程學系
;
Department of Computer Science
In this paper, a new non-iterative clustering method is proposed. It consists of two passes. In the first pass, the mean distance from one object to its nearest neighbor is estimated. Based on this distance, those noises far away from objects are extracted and removed. In the second pass, the mean distance from the remaining objects to their nearest neighbors is computed. Based on the distance, all the intrinsic clusters are then found. The proposed method is non-iterative and can automatically determine the number of clusters. Experimental results also show that the partition generated by the proposed method is more reasonable than that of the well-known c-means algorithm in many complicated object distributions.
A past search algorithm for vector quantization using means and variances of codewords
(1995)
LEE, CH
;
CHEN, LH
;
資訊工程學系
;
Department of Computer Science
FAST CLOSEST CODEWORD SEARCH ALGORITHM FOR VECTOR QUANTIZATION
(1994-06-01)
LEE, CH
;
CHEN, LH
;
交大名義發表
;
資訊工程學系
;
National Chiao Tung University
One of the most serious problems for vector quantisation is the high computational complexity of searching for the closest codeword in the codebook design and encoding phases. The authors present a fast algorithm to search for the closest codeword. The proposed algorithm uses two significant features of a vector, mean value and variance, to reject many unlikely codewords and saves a great deal of computation time. Since the proposed algorithm rejects those codewords that are impossible to be the closest codeword, this algorithm introduces no extra distortion than conventional full search method. The results obtained confirm the effectiveness of the proposed algorithm.
HANDWRITTEN CHARACTER-RECOGNITION USING A 2-LAYER RANDOM GRAPH MODEL BY RELAXATION MATCHING
(1990)
CHEN, LH
;
LIEH, JR
;
交大名義發表
;
資訊工程學系
;
National Chiao Tung University
HIGH-SPEED CLOSEST CODEWORD SEARCH ALGORITHMS FOR VECTOR QUANTIZATION
(1995-05-01)
LEE, CH
;
CHEN, LH
;
資訊工程學系
;
Department of Computer Science
One of the most serious problems for vector quantization is the high computational complexity involved in searching for the closest codeword through a codebook in both codebook design and encoding phases. In this paper, based on the assumption that the distortion is measured by the squared Euclidean distance, two high-speed search methods will be proposed to speed up the search process. The first one uses the difference between the mean values of two vectors to reduce the search space. The second is to find the Karhunen-Loeve transform (KLT) for the distribution of the set of training vectors and then applies the partial distortion elimination method to the transformed vectors. Experimental results show that the proposed methods can reduce lots of mathematical operations.
INTRAMOLECULAR DIELS-ALDER REACTIONS OF FURANS WITH ALLENYL ETHERS - FORMATION OF INDANONE AND TETRALONE
(1991-12-01)
LIN, CC
;
CHEN, LH
;
WU, HJ
;
交大名義發表
;
應用化學系
The intramolecular Diels-Alder reactions of the furans 3, 11a and 11b produced indanone 6 and tetralones 14a and 14b respectively, presumably via the corresponding allenyl ethers 4, 12a and 12b and the intramolecular cycloadducts 5, 13a and 13b.
MOMENT-PRESERVING CURVE DETECTION
(1988-01-01)
CHEN, LH
;
TSAI, WH
;
交大名義發表
;
資訊工程學系
;
National Chiao Tung University
MOMENT-PRESERVING LINE DETECTION
(1988)
CHEN, LH
;
TSAI, WH
;
資訊工程學系
;
Department of Computer Science
MOMENT-PRESERVING SHARPENING - A NEW APPROACH TO DIGITAL PICTURE DEBLURRING
(1988-01-01)
CHEN, LH
;
TSAI, WH
;
資訊工程學系
;
Department of Computer Science
RANDOM-SAMPLING THRESHOLDING - A NEW APPROACH TO MULTILEVEL THRESHOLDING
(1993-12-01)
YIN, PY
;
CHEN, LH
;
資訊工程學系
;
Department of Computer Science
In this paper, a new approach to multilevel thresholding is proposed. First, the approach uniformly spreads some coarse thresholds over the gray-level range. Then, an adjusting process is applied repeatedly until a certain stopping criterion is reached. In the process, some pixels are first randomly sampled. Then, those coarse thresholds are finely adjusted according to the distribution of their neighborhood in the histogram formed by the randomly sampled pixels. The proposed method can automatically determine the number of thresholds. Furthermore, if the number of the resulting thresholds is not sufficient under a certain sense, the proposed method can be iteratively applied, based on the previous results, to get more thresholds. Experimental results also show that the performance of the proposed method is comparable to that of the recent method presented by Lim.