Chen-Chiung Hsieh received his B.S., M.S., and Ph.D. degrees in the Department of Computer Science and Information Engineering, National ChiaoTung University, Hsinchu, Taiwan, in 1986, 1988, and 1992, respectively. From Dec. 1992 to Jan. 2004, he was with the Institute for Information Industry (III) as a vice director. From Dec. 2004 to Jan. 2006, he joined Acer Inc. as a senior director. He is presently a Full Professor in the Department of Computer Science and Engineering at Tatung University, Taipei, Taiwan. His research area is mainly focused in image and multimedia processing.
In order to save manpower on rail track inspection, computer vision-based methodologies are devel... more In order to save manpower on rail track inspection, computer vision-based methodologies are developed. We propose utilizing the YOLOv4-Tiny neural network to identify track defects in real time. There are ten defects covering fasteners, rail surfaces, and sleepers from the upward and six defects about the rail waist from the sideward. The proposed real-time inspection system includes a high-performance notebook, two sports cameras, and three parallel processes. The hardware is mounted on a flat cart running at 30 km/h. The inspection results about the abnormal track components could be queried by defective type, time, and the rail hectometer stake. In the experiments, data augmentation by a Cycle Generative Adversarial Network (GAN) is used to increase the dataset. The number of images is 3800 on the upward and 967 on the sideward. Five object detection neural network models—YOLOv4, YOLOv4-Tiny, YOLOX-Tiny, SSD512, and SSD300—were tested. The YOLOv4-Tiny model with 150 FPS is select...
Journal of Applied Science and Engineering 19(1), Mar 1, 2016
[[abstract]]Super resolution is developed to enhance the resolution of images and various kinds o... more [[abstract]]Super resolution is developed to enhance the resolution of images and various kinds of learning based methods were proposed to magnify a single image. This paper presents a 2D hidden Markov model which could do super resolution by using learned image patch pair database. The image patch pairs store the correspondence relation of high-frequency information between low resolution (LR) patches and high resolution (HR) patches. For each input LR patch, the top five similar LR candidate patches in database are searched to construct a 3D cube which can then be modeled by the proposed 2D hidden Markov model (HMM). A novel 2D Viterbi algorithm is developed to find the optimal LR candidate patches that are the most compatible with each other. The resulting super resolution image could be formed by pasting back the corresponding HR patches from patch pair database according to the positions of found optimal LR patches. By objective comparisons of PSNRs/SSIMs and subjective judgment of the generated super resolution images, the proposed 2D HMM method is superior to the traditional interpolation methods and some existing state-of-the-art methods
Proceedings of 1st International Conference on Image Processing
Vectorization plays an important role for map interpretation in geographic information systems. A... more Vectorization plays an important role for map interpretation in geographic information systems. A new method to convert a raster map into a vector is proposed. A scanner was used to convert the paper source map into a raster representation. Text and graphics in the raster image are first segmented. By connecting the neighboring character blocks, text in different orientations can
IEEE International Conference on Networking, Sensing and Control, 2004
This paper presents a reinforcement-learning approach to a navigation system which allows a goal-... more This paper presents a reinforcement-learning approach to a navigation system which allows a goal-directed mobile robot to incrementally adapt to an unknown environment. Fuzzy rules which map current sensory inputs to appropriate actions are built through the reinforcement learning. Simulation results illustrate the performance of the proposed navigation system. In this paper, ACSNFIS is used as the main network architecture
Proceedings of 13th International Conference on Pattern Recognition, 1996
This paper presents a map interpretation system for automatic extraction of high level informatio... more This paper presents a map interpretation system for automatic extraction of high level information (such as parcels and their attributes) from the scanned images of Chinese cadastral maps. Our map interpretation system consists of three main components: text/graphics separation, parcel extraction, and rotated character recognition. The techniques of text/graphics separation and character recognition are robust to the rotation and writing
Abstract—As computers and networks have been developed vigorously, distance learning could be int... more Abstract—As computers and networks have been developed vigorously, distance learning could be integrated with computer vision techniques for the purpose of better learning effects. In this paper, we developed a distance yoga learning system for people to learn/play through the internet. The main point of the interactive learning system essentially consists in that the gesture performed by player, segmented by computer vision techniques, should possess the same silhouette for a given yoga posture. For better accuracy, the learning score is calculated by matching the distance transformation of the player silhouette with stored standard yoga posture. In the experiments, 23 postures were defined and six persons were invited to do each posture three times. About 86 % of the difference between computer scores and the scores given by a yoga teacher falls within-2.5~2.5. Index Terms—Distance learning, interactive learning, yoga learning, computer vision, distance transformation, posture rec...
2020 International Symposium on Computer, Consumer and Control (IS3C), 2020
Intelligent Customer Service is a very popular application at present. Using the previous user... more Intelligent Customer Service is a very popular application at present. Using the previous user's Q & A records and the existing data in the database, combined with semantic analysis, users can quickly acquire what they want to know. The system takes the Taiwan Disease Control center data as an example, which is proved by experiments that the system can replace the traditional customer service work and can provide users with instant information about diseases.
The study explored the critical needs of external customers and internal customers of Ikari coffe... more The study explored the critical needs of external customers and internal customers of Ikari coffee chain stores in Taiwan from a service design perspective. By using semi-structured interviews, induction questionnaires, KJ mining method, and brainstorming, all of them generated ideas for service design. After compiling, classifying, inductively reasoning and analyzing collected data, this study has identified critical needs. Among them, we found that the atmosphere of the coffee shop such as ambient light and music in the service design is the critical need for customers. We hope customers' emotion is joyful when they enjoy their meal at Ikari coffee chain stores. The aim is to have positive effects on customers' emotion. To achieve this, we adopt Internet of things (IoT) techniques to change both lights and music according to the detected facial expressions of customers. Firstly, customer image is sampled every second to detect face and extract features related to expressio...
2020 IEEE International Conference on Consumer Electronics - Taiwan (ICCE-Taiwan), 2020
To save training efforts, reinforcement learning approach is applied to the autonomous vehicle fo... more To save training efforts, reinforcement learning approach is applied to the autonomous vehicle for obstacle avoidance. Therefore, this study is aimed to let the autonomous vehicle to learn from mistakes and readdress its movement accuracy for collision avoidance in working environment. An enhanced learning method Q-learning is used to record and update the Q values for different movement through a table that the autonomous vehicle can use it to determine how and where to move. The Q table is learned through the deep learning neural network which may encounter innumerable situations from the environments and the different actions performed by the autonomous vehicle. In the experiments, the depth camera is adopted as the input device to be not affected by light intensity and road color. The Q table is ready to use after 9000 epochs or about 3.5 hours training. Let the autonomous vehicle run for 3 minutes at a time in three different environments with lights on and off 10 times each. T...
In order to save manpower on rail track inspection, computer vision-based methodologies are devel... more In order to save manpower on rail track inspection, computer vision-based methodologies are developed. We propose utilizing the YOLOv4-Tiny neural network to identify track defects in real time. There are ten defects covering fasteners, rail surfaces, and sleepers from the upward and six defects about the rail waist from the sideward. The proposed real-time inspection system includes a high-performance notebook, two sports cameras, and three parallel processes. The hardware is mounted on a flat cart running at 30 km/h. The inspection results about the abnormal track components could be queried by defective type, time, and the rail hectometer stake. In the experiments, data augmentation by a Cycle Generative Adversarial Network (GAN) is used to increase the dataset. The number of images is 3800 on the upward and 967 on the sideward. Five object detection neural network models—YOLOv4, YOLOv4-Tiny, YOLOX-Tiny, SSD512, and SSD300—were tested. The YOLOv4-Tiny model with 150 FPS is select...
Journal of Applied Science and Engineering 19(1), Mar 1, 2016
[[abstract]]Super resolution is developed to enhance the resolution of images and various kinds o... more [[abstract]]Super resolution is developed to enhance the resolution of images and various kinds of learning based methods were proposed to magnify a single image. This paper presents a 2D hidden Markov model which could do super resolution by using learned image patch pair database. The image patch pairs store the correspondence relation of high-frequency information between low resolution (LR) patches and high resolution (HR) patches. For each input LR patch, the top five similar LR candidate patches in database are searched to construct a 3D cube which can then be modeled by the proposed 2D hidden Markov model (HMM). A novel 2D Viterbi algorithm is developed to find the optimal LR candidate patches that are the most compatible with each other. The resulting super resolution image could be formed by pasting back the corresponding HR patches from patch pair database according to the positions of found optimal LR patches. By objective comparisons of PSNRs/SSIMs and subjective judgment of the generated super resolution images, the proposed 2D HMM method is superior to the traditional interpolation methods and some existing state-of-the-art methods
Proceedings of 1st International Conference on Image Processing
Vectorization plays an important role for map interpretation in geographic information systems. A... more Vectorization plays an important role for map interpretation in geographic information systems. A new method to convert a raster map into a vector is proposed. A scanner was used to convert the paper source map into a raster representation. Text and graphics in the raster image are first segmented. By connecting the neighboring character blocks, text in different orientations can
IEEE International Conference on Networking, Sensing and Control, 2004
This paper presents a reinforcement-learning approach to a navigation system which allows a goal-... more This paper presents a reinforcement-learning approach to a navigation system which allows a goal-directed mobile robot to incrementally adapt to an unknown environment. Fuzzy rules which map current sensory inputs to appropriate actions are built through the reinforcement learning. Simulation results illustrate the performance of the proposed navigation system. In this paper, ACSNFIS is used as the main network architecture
Proceedings of 13th International Conference on Pattern Recognition, 1996
This paper presents a map interpretation system for automatic extraction of high level informatio... more This paper presents a map interpretation system for automatic extraction of high level information (such as parcels and their attributes) from the scanned images of Chinese cadastral maps. Our map interpretation system consists of three main components: text/graphics separation, parcel extraction, and rotated character recognition. The techniques of text/graphics separation and character recognition are robust to the rotation and writing
Abstract—As computers and networks have been developed vigorously, distance learning could be int... more Abstract—As computers and networks have been developed vigorously, distance learning could be integrated with computer vision techniques for the purpose of better learning effects. In this paper, we developed a distance yoga learning system for people to learn/play through the internet. The main point of the interactive learning system essentially consists in that the gesture performed by player, segmented by computer vision techniques, should possess the same silhouette for a given yoga posture. For better accuracy, the learning score is calculated by matching the distance transformation of the player silhouette with stored standard yoga posture. In the experiments, 23 postures were defined and six persons were invited to do each posture three times. About 86 % of the difference between computer scores and the scores given by a yoga teacher falls within-2.5~2.5. Index Terms—Distance learning, interactive learning, yoga learning, computer vision, distance transformation, posture rec...
2020 International Symposium on Computer, Consumer and Control (IS3C), 2020
Intelligent Customer Service is a very popular application at present. Using the previous user... more Intelligent Customer Service is a very popular application at present. Using the previous user's Q & A records and the existing data in the database, combined with semantic analysis, users can quickly acquire what they want to know. The system takes the Taiwan Disease Control center data as an example, which is proved by experiments that the system can replace the traditional customer service work and can provide users with instant information about diseases.
The study explored the critical needs of external customers and internal customers of Ikari coffe... more The study explored the critical needs of external customers and internal customers of Ikari coffee chain stores in Taiwan from a service design perspective. By using semi-structured interviews, induction questionnaires, KJ mining method, and brainstorming, all of them generated ideas for service design. After compiling, classifying, inductively reasoning and analyzing collected data, this study has identified critical needs. Among them, we found that the atmosphere of the coffee shop such as ambient light and music in the service design is the critical need for customers. We hope customers' emotion is joyful when they enjoy their meal at Ikari coffee chain stores. The aim is to have positive effects on customers' emotion. To achieve this, we adopt Internet of things (IoT) techniques to change both lights and music according to the detected facial expressions of customers. Firstly, customer image is sampled every second to detect face and extract features related to expressio...
2020 IEEE International Conference on Consumer Electronics - Taiwan (ICCE-Taiwan), 2020
To save training efforts, reinforcement learning approach is applied to the autonomous vehicle fo... more To save training efforts, reinforcement learning approach is applied to the autonomous vehicle for obstacle avoidance. Therefore, this study is aimed to let the autonomous vehicle to learn from mistakes and readdress its movement accuracy for collision avoidance in working environment. An enhanced learning method Q-learning is used to record and update the Q values for different movement through a table that the autonomous vehicle can use it to determine how and where to move. The Q table is learned through the deep learning neural network which may encounter innumerable situations from the environments and the different actions performed by the autonomous vehicle. In the experiments, the depth camera is adopted as the input device to be not affected by light intensity and road color. The Q table is ready to use after 9000 epochs or about 3.5 hours training. Let the autonomous vehicle run for 3 minutes at a time in three different environments with lights on and off 10 times each. T...
Uploads
Papers by Chen-Chiung Hsieh