Browsing by Author "LEE, HJ"

Now showing 1 - 17 of 17

3-FRAME CORNER MATCHING AND MOVING OBJECT EXTRACTION IN A SEQUENCE OF IMAGES
(1990-11-01)
LEE, HJ
;
DENG, HC
;
資訊工程學系
;
Department of Computer Science
A HIERARCHICAL MODEL-GUIDED GENERATION OF CHINESE CHARACTERS
(1994)
LEE, HJ
;
HSU, HC
;
資訊工程學系
;
Department of Computer Science
A PROBABILISTIC STROKE-BASED VITERBI ALGORITHM FOR HANDWRITTEN CHINESE CHARACTERS RECOGNITION
(1992)
HSIEH, CC
;
LEE, HJ
;
資訊工程學系
;
Department of Computer Science
ADAPTIVE TRACKING SYSTEM FOR A MANEUVERING TARGET USING IMAGES WITH CORRELATED NOISES
(1989-05-01)
CHANG, KC
;
LEE, HJ
;
CHUNG, CG
;
交大名義發表
;
資訊工程學系
DETERMINATION OF 3D HUMAN-BODY POSTURES FROM A SINGLE VIEW
(1985)
LEE, HJ
;
ZEN, C
;
交大名義發表
;
資訊工程學系
;
National Chiao Tung University
ENHANCED DIAMAGNETISM IN YBA2CU3O7-DELTA BY HYDROGEN-PLASMA TREATMENT
(1994-09-10)
LEE, MW
;
TAI, MF
;
LEE, HJ
;
HWANG, HL
;
KAO, HCI
We report that the diamagnetism of as-grown YBa2Cu3O7-delta powders can be enhanced by treatment with RF-generated hydrogen plasma. The field-cooling susceptibility at 5 K is found to increase by up to 24%. The enhancement is reproducible for all five samples studied. The lattice parameter c decreases after hydrogenation. For a sample treated for the shortest period, its pinning potential, obtained from magnetic relaxation measurements, is reduced following H-treatment.
INCREASING CHARACTER-RECOGNITION ACCURACY BY DETECTION AND CORRECTION OF ERRONEOUSLY IDENTIFIED CHARACTERS
(1994-09-01)
TUNG, CH
;
LEE, HJ
;
資訊工程學系
;
Department of Computer Science
This paper presents a general approach to increasing character recognition accuracy by detection and correction of erroneously identified characters. A high-accuracy two-stage recognition system for recognizing handwritten Chinese text with 5401 categories is demonstrated. In the first stage of the system, two matching modules are applied to recognize an input character simultaneously. The first matching module uses a direction feature and the second uses a generalized feature. A character is rejected at the first stage when the matching results of the two modules are not the same. Because our first stage recognizes most of the input characters correctly and outputs a small number of candidates for each rejected character, a bigram Markov language model in the second stage can choose a candidate with high recognition rate for each rejected character according to contextual information. Experiments are performed on sentences consisting of characters extracted from the CCL/HCCR1 handwritten Chinese character database. In the first stage, the rejection rate for the input characters is 13% and the recognition rate for the accepted characters is 95.9%. In the second stage, the recognition rate for the characters rejected by the first stage is 91.2%. Thus, the overall recognition rate for the input handwritten text is 95.9% x 0.87 + 91.2% x 0.13 (= 95.2%)
KNOWLEDGE-GUIDED VISUAL-PERCEPTION OF 3-D HUMAN GAIT FROM A SINGLE IMAGE SEQUENCE
(1992-03-01)
CHEN, Z
;
LEE, HJ
;
交大名義發表
;
資訊工程學系
;
National Chiao Tung University
A computer vision method is presented to determine the 3-D spatial locations of joints or feature points of a human body from a film recording the human motion during walking. The proposed method first applies the geometric projection theory to obtain a set of feasible postures from a single image, then it makes use of the given dimensions of the human stick figure, physiological and motion-specific knowledge to constrain the feasible postures in both the single-frame analysis and the multi-frame analysis. Finally a unique gait interpretation is selected by an optimization algorithm. Computer simulations are used to illustrate the ideas presented.
LINE-BASED STRUCTURAL MATCHING VIA SEGMENT SPLITTING
(1990-03-01)
LEE, HJ
;
YU, DJ
;
交大名義發表
;
資訊工程學系
;
National Chiao Tung University
MANEUVERING TARGET TRACKING IN A CLUTTERED ENVIRONMENT FROM NOISY IMAGES BY AN ADAPTIVE PROBABILISTIC DATA ASSOCIATION FILTER
(1991-01-01)
LEE, HJ
;
CHUNG, CG
;
CHANG, KC
;
交大名義發表
;
資訊工程學系
A new tracking method for estimating motion parameters of a manoeuvring target in a cluttered environment from noisy image sequences is presented. Traditionally, tracking systems have been broadly classified according to (1) moving target with or without manoeuvring, and (2) tracking in a clean (uncluttered) or cluttered environment. The difficulties studied here are those arising from (a) unknown acceleration inducing a randomly updated trajectory, and (b) the disturbance of returns from false alarms or other interfering objects. The probability data association filter (PDAF) is augmented by the semi-Markov process, called the adaptive PDAF, to handle the difficulties. Since the acceleration state applied either by a pilot or by a guidance control system is independent of the appearance of false returns, the treatment of target manoeuvring and false returns are separated into two sub-problems, i.e. the PDAF is applied to govern the disturbance of multiple returns and the semi-Markov process is applied to estimate uncertain accelerations. Three simulation results of variations show small biases from the true trajectory. By comparing with a nearest neighbour standard filter, the proposed tracking system also demonstrates better results.
MULTIFRAME SHIP DETECTION AND TRACKING IN AN INFRARED IMAGE SEQUENCE
(1990)
LEE, HJ
;
HUANG, LF
;
CHEN, Z
;
交大名義發表
;
資訊工程學系
MULTISTAGE PRE-CANDIDATE SELECTION IN HANDWRITTEN CHINESE CHARACTER-RECOGNITION SYSTEMS
(1994-08-01)
TUNG, CH
;
LEE, HJ
;
TSAI, JY
;
資訊工程學系
;
Department of Computer Science
This paper presents a model for reducing the processing time needed for recognition of handwritten Chinese characters via multi-stage pre-candidate selection. The features used in pre-candidate selection are ordered according to the reduction rates evaluated from a set of training characters. A multi-stage pre-candidate selection module is activated for each input character. In each stage, we use a single feature computed from the input character to eliminate impossible categories of characters. When the number of stages increases, the pre-candidate selection time will increase, while the candidate selection time will decrease. The total execution time will decrease when the increase in pre-candidate selection time is less than the decrease in candidate selection time. The total execution time can be predicted from the set of training characters. The number of stages of the pre-candidate selection module is selected to be that which results in the minimum total execution time. Experiments are performed on the CCL/HCCR1 database with 5401 categories of handwritten Chinese characters. The experimental results show that the proposed model can reduce the total execution time significantly without decreasing the precision of the candidate selection module.
OFF-LINE RECOGNITION OF HANDWRITTEN CHINESE CHARACTERS BY ONLINE MODEL-GUIDED MATCHING
(1992-11-01)
HSIEH, CC
;
LEE, HJ
;
資訊工程學系
;
Department of Computer Science
A model-guided structural matching method for recognizing handwritten off-line Chinese characters is presented. According to the stroke writing sequence, each character is described by an on-line model, which is a one-dimensional (1D) string consisting of a stroke sequence interleaved with relationships between two consecutive strokes. After an unknown input is thinned and line approximated, all possible strokes are extracted. A tree is created to represent all possible combinations of the extracted strokes which have the same stroke sequence as the one defined in the model. In the recognition process, we match the strokes with those defined in the on-line model. The matching process is formulated as a tree searching algorithm guided by the relationships in the model. The modelled character is taken to be a candidate if there is a feasible path which satisfies the relationships defined in the on-line model and the number of missing strokes in the input is less than a given threshold. The input is recognized as the character with the greatest number of matched strokes among all candidates. Experimental results on 300 frequently used characters in a database called CCL/HCCR1, which contains 5401 Chinese characters and about 250 variations for each one, show that the recognition rate is over 90%.
PERFORMANCE ANALYSIS OF AN OCR SYSTEM VIA AN ARTIFICIAL HANDWRITTEN CHINESE CHARACTER GENERATOR
(1994-02-01)
TUNG, CH
;
CHEN, YJ
;
LEE, HJ
;
資訊工程學系
;
Department of Computer Science
A handwritten Chinese character generator that generates handwritten Chinese characters with different variations is proposed and used to evaluate the performance of an OCR system. The character generator first generates radicals of a character stroke by stroke and then combines the generated radicals to form a line-vector character. Next, the line-vector character is thickened to obtain a character image. Characters generated by this method will have great variance in shape but will still satisfy the structural constraints. The generated characters are then used to perform two types of evaluations. First, the stability of stroke extractors in terms of stroke number is evaluated. Two stroke extractors, one based on thinning and the other on vectorization, are analyzed. The vectorization-based stroke extractor we propose operates directly on the run length codes of line segments. From experimental results, we find that the thinning-based method is more time-consuming but more stable than the vectorization-based method. Second, the peak performance of a matching module is evaluated and the recognition error caused by stroke extractors is identified.
RECOGNITION OF HANDWRITTEN CHINESE CHARACTERS VIA SHORT LINE SEGMENTS
(1992-05-01)
LEE, HJ
;
CHEN, B
;
交大名義發表
;
資訊工程學系
;
National Chiao Tung University
A system to recognize handwritten Chinese characters is presented. In the first stage, short line segments are extracted, where a new efficient algorithm is proPosed, based on accumulated chain codes, for line approximation. In the feature extraction stage, features of each character are computed, which are further defined by the features of its composing line segments. In the matching stage, dynamic programming is first used to calculate the similarity between a segment of the input character and a segment of the reference character, and then the similarity between the two characters is computed. In order to reduce the number of candidates, a coarse classification algorithm is proposed. The recognition rate for 150 characters is about 90%, excluding the error in coarse classification.
REGION MATCHING AND DEPTH FINDING FOR 3D OBJECTS IN STEREO AERIAL PHOTOGRAPHS
(1990)
LEE, HJ
;
LEI, WL
;
交大名義發表
;
資訊工程學系
;
National Chiao Tung University
UNDERSTANDING MATHEMATICAL EXPRESSIONS USING PROCEDURE-ORIENTED TRANSFORMATION
(1994-03-01)
LEE, HJ
;
LEE, MC
;
交大名義發表
;
資訊工程學系
;
National Chiao Tung University
A system for understanding mathematical expressions is presented. The mathematical expressions scanned from a printed document are recognized and transformed into one-dimensional (1D) strings according to the format of a publication system. After separating all symbols in an input mathematical expression, 13 features are utilized to represent each symbol. In order to reduce the computational time, a coarse classification algorithm is applied to reduce the number of candidates. Then for each input symbol, the character with the highest similarity is selected as the candidate symbol. Since some of the symbols in an arithmetical expression may touch each other, a dynamic programming algorithm which uses structural features is adopted to identify correct characters from connected symbols. In the expression formation stage, a procedure-oriented method is proposed to translate the recognized symbols from a two-dimensional (2D) space into a 1D character string. A total of 105 mathematical expressions have been used as training data and 50 expressions as testing data. The experimental results have demonstrated the feasibility of the understanding system.