Browsing by Author "Lin, Guan-Ting"

Now showing 1 - 6 of 6

A Cost-Effective Planning Graph Approach for Large-Scale Web Service Composition
(2012)
Lin, Szu-Yin
;
Lin, Guan-Ting
;
Chao, Kuo-Ming
;
Lo, Chi-Chun
;
資訊管理與財務金融系註：原資管所+財金所
Web Service Composition (WSC) problems can be considered as a service matching problem, which means that the output parameters of a Web service can be used as inputs of another one. However, when a very large number of Web services are deployed in the environment, the service composition has become sophisticated and complicated process. In this study, we proposed a novel cost-effective Web service composition mechanism. It utilizes planning graph based on backward search algorithm to find multiple feasible solutions and recommends a best composition solution according to the lowest service cost. In other words, the proposed approach is a goal-driven mechanism, which can recommend the approximate solutions, but it consumes fewer amounts of Web services and less nested levels of composite service. Finally, we implement a simulation platform to validate the proposed cost-effective planning graph mechanism in large-scale Web services environment. The simulation results show that our proposed algorithm based on the backward planning graph has reduced by 94% service cost in three different environments of service composition that is compared with other existing service composition approaches which are based on a forward planning graph.
Efficient and Lightweight Convolutional Neural Network for Lane Mark and Road Segmentation
(2018-01-01)
Lin, Guan-Ting
;
Guo, Jiun-In
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
Semantic segmentation is one of an important task in computer vision that takes a great part in the perception needs of intelligent autonomous vehicles. ConvNets excel at this task, as they can be adaptively trained end-to-end to yield a set of robust hierarchies of features. The proposed key method is to reduce the unnecessary weights to build an efficient and lightweight network to acquire high accuracy on lane mark and road segmentation at pixel level. The proposed fully convolutional neural network achieves 360x480@28 fps and 97.6% accuracy on our in-house pixel-based hand-annotated lane mark and road datasets. All our models and results are trained and evaluated on an NVIDIA GTX 1080 GPU device.
Learning with Detail and Morphological Refinement for Satellite Image Analysis based on Convolutional Neural Network
(2018-01-01)
Lin, Guan-Ting
;
Yang, Yung-I
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
we present a multiple stages way to analyze the content of satellite images. Our methodology is divided into three major steps. First, Convolutional Neural Networks (CNNs) for semantic segmentation between buildings and nature scene. Then, the output semantic image would be refined in morphology way. In the last stage, Depth-First Search (DFS) algorithm is used for buildings counting. The experimental results show that refined images have smother boundaries. Base on the output images, we can count buildings using DFS algorithm accurately by refined image.
One Stage Detection Network with an Auxiliary Classifier for Real-Time Road Marks Detection
(2018-01-01)
Lin, Guan-Ting
;
Santoso, Patrisia Sherryl
;
Lin, Che-Tsung
;
Tsai, Chia-Chi
;
Guo, Jiun-In
We construct a robust road mark detector that achieves high accuracy with real-time processing performance ( 32 fps) under nVidia Titan-X GPU. We combine one stage deep learning detector with auxiliary CNN classifiers as a robust road marks detector. We found out that one stage detector not only detects multiple objects via single inference efficiently, but also remains a good accuracy in performance perspective. However, to make it better, we add an extra CNN classifier as the back part of the proposed architecture to reduce false positive and get better accuracy. The proposed detector can achieve 86.8% mAP in our in-house six-class road mark database.
RECOGNIZING CHINESE TEXTS WITH 3D CONVOLUTIONAL NEURAL NETWORK
(2019-01-01)
Chen, Kuan-Chou
;
Lin, Guan-Ting
;
Lin, Che-Tsung
;
Guo, Jiun-In
;
交大名義發表
In this paper, we propose a deep learning system to localize and recognize Chinese texts in scenes with signage and road marks through 3D convolutional neural network. The proposed system adopts YOLO for detecting target location and exploits 3D convolutional neural network for recognizing the contents. The proposed design outperforms the existing designs based on LSTM and achieves real-time processing performance, which is feasible to be implemented on embedded platforms. The proposed system reaches over 90% accuracy in recognizing Chinese texts on bird's-eye viewing road marks in a self-driving vehicle equipped with a fisheye camera. In addition, this system can achieve 20 fps execution speed with NVIDIA DIGITS DevBox with 1080Ti GPU, which is fast enough for autonomous driving applications.
Stop Line Detection and Distance Measurement for Road Intersection based on. Deep Learning Neural Network
(2017-01-01)
Lin, Guan-Ting
;
Santoso, Patrisia Sherryl
;
Lin, Che-Tsung
;
Tsai, Chia-Chi
;
Guo, Jiun-In
In this paper, we introduce Boost-CNN, a robust stop-line detector that can detect objects (stop line) with competitive tradeoff between speed and accuracy. Boost-CNN consists of an AdaBoost classifier and a CNN. The former is our region proposal generator and it is further combined with the later to be a stop-line detector. In addition, an automatic hard mining method is proposed to reduce the number of false alarm. Our proposed detector achieves 91.5% in accuracy and has 100 FPS performance in test time (performed on NVITAA DIGITS DevBox and Titan X GPU).