The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
It is important to extract pattern's feaure and recognize them by proper model in pattern classification. In this paper, we propose the modified LDA to compress the extracted feature, and design the classifier with Precision Constrained Gaussian Model. A series of experiments are offered and the experimental result shows that our PCGM can achieve a much better generalization and MLDA has a better...
The increasing volume of spam has become a serious threat not only to the Internet, but also to the society. However, it's a great challenge to discover the spam from the Internet effectively and efficiently. Content-based filtering is one of the mainstream methods to solve the problem. This paper proposed a content based spam topic detection strategy through keyword extraction. In particular, spam...
A method to locate the axis of radio frequency ablation electrode(RFAE) in 3D Ultrasound(US) image is presented based on 3D phase-grouping in this paper. Firstly, all voxels in 3D US images are categorized into different groups which are called Line Support Region(LSR) according to the outer products of adjacent orientation vectors. And then, the RFAE axis is extracted with 3D Randomized Hough transform...
Through research on the calculation method of feature words' weight in texts and semantic similarity between words, we proposed a calculation method of feature words' weight based on concept weight for the semantic association phenomenon of text features and the prevalence of high-dimensional problem in a text vector space model. This method reduces the semantic loss of the feature set and the dimension...
Image perceptual hashing find applications in content indexing, large-scale image database management, certification and authentication and digital watermarking. We propose a Block-DCT and PCA based image perceptual hash in this article and explore the algorithm in the application of tamper detection. The main idea of the algorithm is to integrate color histogram and DCT coefficients of image blocks...
An obstacle system based on visual selective feature and stereo vision is stated in this paper. By extracting feature vectors with color, intensity and direction, the corresponding saliency map is constructed. Then, obstacle area is figured out through threshold segmentation to the saliency map, and 3-D information of the obstacles is computed by using stereo vision. Finally, obstacle information...
An image watermarking scheme resisting RST (rotation, scaling, translation) based on RST invariant square region is proposed in the paper. Two SIFT (scale-invariant feature transform) feature points are selected to determine the position, size, direction of the region. The watermark embedding and detection is completed in the square region with quantization, which could guarantee blind detection....
In Computer Vision, problem of identifying or classifying the objects present in an image is called Object Categorization. It is challenging problem, especially when the images have clutter background, occlusions or different lighting conditions. Many vision features have been proposed which aid object categorization even in such adverse conditions. Past research has shown that, employing multiple...
Much attention has been paid to WLAN indoor positioning algorithm for its high accuracy and low cost to meet the location based services (LBS). This paper proposes a novel positioning algorithm based on positioning characteristics extraction in WLAN indoor environment. Each RSS signal from an individual access point is taken as input of the RBF neural networks to establish the mapping between RSS...
This paper presents an automatic people counting system based on face detection, where the number of people passing through a gate or door is counted by setting a video camera. The basic idea is to first use the frame difference to detect the rough edges of moving people and then use the chromatic feature to locate the people face. Based on NCC (Normalized Color Coordinates) color space, the initial...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.