Fanhuai Shi
Shanghai Jiao Tong University
Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by Fanhuai Shi.
Engineering Applications of Artificial Intelligence | 2009
Xixia Huang; Fanhuai Shi; Wei Gu; Shanben Chen
This paper proposes a support vector machine-based fuzzy rules acquisition system (SVM-FRAS) for modeling of the gas tungsten arc welding (GTAW) process. The character of SVM in extracting support vector provides a mechanism to extract fuzzy IF-THEN rules from the training data set. We construct the fuzzy inference system using fuzzy basis function. The gradient technique is used to tune the fuzzy rules and the inference system. Theoretical analysis and comparative tests are performed comparing with other fuzzy systems. Modeling is one of the key techniques in the automatic control of the arc welding process, and is still a very difficult problem. Comprehensibility is one of the required characteristics in modeling for the complex GTAW process. We use the proposed SVM-FRAS to obtain the rule-based model of the aluminum alloy pulse GTAW process. Experimental results show the SVM-FRAS model possesses good generalization capability as well as high comprehensibility.
international symposium on visual computing | 2009
Fanhuai Shi; Xixia Huang; Ye Duan
This paper proposes a robust Harris-Laplace detector by scale multiplication. The specific Harris corner measure functions at adjacent scales are multiplied as a product function to magnify the corner like structures, while suppress the image noise and weak features simultaneously. Unlike the contour-based multi-scale curvature product for image corner detection, we detect the corner like features directly in intensity image. Experiments on natural images demonstrate that the proposed method has good consistency of corner detection under different noise levels.
Journal of Intelligent and Robotic Systems | 2011
Fanhuai Shi; Yongjian Xi; Xiaoling Li; Ye Duan
This paper presents a prototype system of rooftop detection and 3D building modeling from aerial images. In this system, without the knowledge of the position and orientation information of the aerial vehicle a priori, the parameters of the camera pose and ground plane are first estimated by simple human–computer interaction. Next, after an over-segmentation of the aerial image by the Mean-Shift algorithm, the rooftop regions are coarsely detected by integrating multi-scale SIFT-like feature vectors with SVM-based visual object recognition. 2D cues alone however might not always be sufficient to separate regions such as parking lots from building roofs. Thus in order to further refine the accuracy of the roof-detection result and remove the misclassified non-rooftop regions such as parking lots, we further resort to 3D depth information estimated based on multi-view geometry. More specifically, we determine whether a candidate region is a rooftop or not according to its height information relative to the ground plane, whereas the candidate region’s height information is obtained by a novel, hierarchical, asymmetry correlation-based corner matching scheme. The output of the system will be a water-tight triangle mesh based 3D building model texture mapped with the aerial images. We developed an interactive 3D viewer based on OpenGL and C+ + to allow the user to virtually navigate the reconstructed 3D scene with mouse and keyboard. Experimental results are shown on real aerial scenes.
Archive | 2011
Fanhuai Shi; Xixia Huang; Ye Duan
Robust and high accuracy corner matching plays an essential role in many applications in computer vision such as camera calibration, 3D reconstruction and robot localization. In this paper, we describe a hybrid approach that can automatically detect and match image corners with high accuracy. Our approach is based on SIFT structure information and sub-pixel Harris corner localization, which is rotation invariant and is localized directly on true image corners detected by the enhanced curvature scale space method. Experimental results show that the proposed method offers an effective solution to automatic robust corner matching.
Lecture Notes in Control and Information Sciences | 2007
Xixia Huang; Fanhuai Shi; Shanben Chen
This paper proposes a support vector machine (SVM)-based fuzzy system (SVM-FS), which has good comprehensibility as well as satisfactory generalization capability. SVM provides a mechanism to extract support vectors for generating fuzzy IF-THEN rules from training data. In SVM-FS, SVM is used to extract IF-THEN rules; the fuzzy basis function inference system is adopted as the fuzzy inference system. Furthermore, we theoretically analyze the proposed SVM-FS on the rule extraction and the inference method comparing with other fuzzy systems; comparative tests are performed using benchmark data. The analysis and the experimental results show that the new approach possesses high comprehensibility as well as satisfactory generalization capability
international symposium on visual computing | 2009
Fanhuai Shi; Yongjian Xi; Xiaoling Li; Ye Duan
This paper presents a new procedure for rooftop detection and 3D building modeling from aerial images. After an over-segmentation of the aerial image, the rooftop regions are coarsely detected by employing multi-scale SIFT-like features and visual object recognition. In order to refine the detected result and remove the non-rooftop regions, we further resort to explore the 3D information of the rooftop by 3D reconstruction. Wherein, we employ a hierarchical strategy to obtain the corner correspondence between images based on an asymmetry correlation corner matching. We determine whether a candidate region is a rooftop or not according to its height information relative to the ground plane. Finally, the 3D building model with texture mapping based on one of the images is given. Experimental results are shown on real aerial scenes.
ieee conference on cybernetics and intelligent systems | 2008
Xixia Huang; Jiansheng Pan; Fanhuai Shi; Shanben Chen
This paper investigates both the knowledge model and the mechanism model for correcting carbon potential using an oxygen sensor (CPUOS). CPUOS is widely used and there exists a deviation between the true value and the measured value. Therefore it is very important to study the correction model for CPUOS. Experiments are planned and carried out to generate the necessary data. Based on the experimental data we get the knowledge model for CPUOS using support vector machine (SVM). Under the guidance of the knowledge model, we build the mechanism model based on the carbon potential relevant theory. The knowledge model and the mechanism model are corrected and verified by the practical experience.
International Heat Treatment & Surface Engineering | 2008
Xixia Huang; Jiansheng Pan; Chujun Qian; Fanhuai Shi; Shanben Chen
AbstractKnowledge based and mechanistic models for correcting carbon potential using an oxygen sensor (CPUOS) are considered. Although CPUOS is widely used, there usually exists a deviation between the true and measured values. Therefore it is important to establish a valid correction model. Experiments were designed and carried out to generate data for modelling and to identify the main influencing factors. Both single and multivariable knowledge models have been established using the support vector machine approach. Guided by these models, a mechanistic model based on the relevant theory of carbon potential has been built up.
The International Journal of Advanced Manufacturing Technology | 2003
Fanhuai Shi; Z.L. Lou; Yuelong Zhang; J.G. Lu
The International Journal of Advanced Manufacturing Technology | 2003
Fanhuai Shi; Z.L. Lou; Yuelong Zhang; J.G. Lu