Titlebook: Artificial Intelligence and Computational Intelligence; Second International Hepu Deng,Duoqian Miao,Fu Lee Wang Conference proceedings 2011

intercede

減少

Wernickes-area

An Object Recognition Strategy Base upon Foreground Detectionble information for us. Object recognition is necessary in processing video information. However, traditional recognition methods based on object segmentation can hardly meet the system demands for running online. In this paper, we have made use of SVM based upon HOG feature descriptors to achieve o

MAL

Transfusion

Lane Detection in Critical Shadow Conditions Based on Double A/D Convertors Camera proposed. Based on the color information of the unstructured lane, an improved region-growing algorithm is employed to segment the image. To enhance the image quality and the accuracy of the algorithm, a double A/D convertors camera is used to recover the color space information of the environments

dissolution

Image Selection Based on Grayscale Features in Robotic Weldingmages obtained in the welding process because it is easily affected by spattering, fuming and electromagnetic noise. The paper introduces a method to select useful images before further processing. Experimental tests are conducted to verify its accuracy.

禁止

騷動

1FAWN

Fabric Defect Detection Based on Computer Visionted by the subjective factors of inspectors, an automatic computer vision based fabric defect detection method is introduced in this paper. The system uses threshold segmentation method to identify if there are any defects existed in the fabric, adopts image feature based approach to recognize oil s

綁架