Multiple instance learning (MIL) is a form of weakly supervised learning for problems in which training instances are arranged into bags, and a label is provided for whole bags but not for individual instances. Most p...
详细信息
ISBN:
(纸本)9781467389105
Multiple instance learning (MIL) is a form of weakly supervised learning for problems in which training instances are arranged into bags, and a label is provided for whole bags but not for individual instances. Most proposed MIL algorithms focus on bag classification, but more recently, the classification of individual instances has attracted the attention of the patternrecognition community. While these two tasks are similar, there are important differences in the consequences of instance misclassification. In this paper, the scoring function learned by MIL classifiers for the bag classification task is exploited for instance classification by adjusting the decision threshold. A new criterion for the threshold adjustment is proposed and validated using 7 reference MIL algorithms on 3 real-world data sets from different application domains. Experiments show considerable improvements in accuracy over these algorithms for instance classification. In some applications, the unweighted average recall increases by as much as 18%, while the F-1-score increases by 12%.
The problem of emotion state recognition using the sigma-pi artificial neural network is considered. The specific feature of the considered network is the presence of two different types of activation functions: sigmo...
详细信息
ISBN:
(纸本)9781509037360
The problem of emotion state recognition using the sigma-pi artificial neural network is considered. The specific feature of the considered network is the presence of two different types of activation functions: sigmoid and bell-shaped. A learning algorithm for the sigma-pi network is proposed. This algorithm is characterized by high approximation accuracy especially for nonlinear processes in real time. Training and testing network on image dataset was made.
In ISO TC249 conference, tongue diagnosis has been one of the most active research and their objectifications has become significant with the help of numerous statistical and machinelearning algorithm. Color informat...
详细信息
ISBN:
(数字)9781510604315
ISBN:
(纸本)9781510604315
In ISO TC249 conference, tongue diagnosis has been one of the most active research and their objectifications has become significant with the help of numerous statistical and machinelearning algorithm. Color information of substance or tongue body has kept valuable information regarding the state of disease and its correlation with the internal organs. In order to produce high reproducibility of color measurement analysis, tongue images have to undergo several procedures such as color correction, segmentation and tongue's substance-coating separation. This paper presents a novel method to recognize substance and coating from tongue images and eliminate the tongue coating for accurate substance color measurement for diagnosis. By utilizing Hue, Saturation, Value (HSV) color space, new color-brightness threshold parameters have been devised to improve the efficiency of tongue's substance and coating separation procedures and eliminate shadows. The algorithm offers fastprocessing time around 0.98 seconds for 60,000 pixels tongue image. The successful tongue's substance and coating separation rate reported is 90% compared to the labelled data verified by the practitioners. Using 300 tongue images, the substance Lab color measurement with small standard deviation had revealed the effectiveness of this proposed method in computerized tongue diagnosis system.
Object recognition and texture identification are the two main application of sonar imageprocessing. It consists of 3 main steps such as image segmentation, object recognition and pattern identification. Before all t...
详细信息
ISBN:
(纸本)9781467367257
Object recognition and texture identification are the two main application of sonar imageprocessing. It consists of 3 main steps such as image segmentation, object recognition and pattern identification. Before all these steps, some of the preprocessing procedure are included such as filtering of noise, edge detection etc. Many edge detection operators are available and also a mathematical modeling called morphological processing is also very useful for edge detection. Color images generally have more visual perception over gray scale and monochrome images. Therefore in this paper the standard edge detection techniques along with the morphological processing are carried out. Comparison of the result along with the pseudo colored image proves that the morphological processing along with edge detection techniques gives more information than another one. Subjective analysis tells that, this approach mainly useful for the sonar image objects consist of larger shadow regions.
The proceedings contain 108 papers. The special focus in this conference is on Smart Trends in Information Technology and Computer Communications. The topics include: Real time sign language processing system;nano sca...
ISBN:
(纸本)9789811034329
The proceedings contain 108 papers. The special focus in this conference is on Smart Trends in Information Technology and Computer Communications. The topics include: Real time sign language processing system;nano scale dual material gate silicon on nothing junctionless MOSFET for improving short channel effect and analog performance;an improved image compression technique using Huffman coding and FFT;comparison and analysis of cuckoo search and firefly algorithm for image enhancement;an exploration of miscellaneous palm print recognition modalities;plugin for instantaneous web page rejuvenation and translation;issues and requirements for successful integration of semantic knowledge in web usage mining for effective personalization;image fusion based on the modified curvelet transform;a case study of Indian hospitality and tourism sector;a comparative study of various algorithms in wireless networking;research and analysis of open security issues in communication for wireless sensor network;empirical analysis of image segmentation techniques;sentiment analysis at document level;an approach to sentiment analysis on unstructured data in big data environment;lung cancer diagnosis by hybrid support vector machine;multi chromatic balls with relaxed criterion to detect larger communities in social networks;image segmentation and object recognition using machinelearning;proposed algorithms to the state explosion problem;predicting software maintainability using object oriented dynamic complexity measures;security enhancement of blowfish block cipher and comparative analytical study for news text classification techniques applied for stock market price extrapolation.
QR codes have become useful and efficient data storage tools which are exploited in many commercial applications including product tracking, website redirection, etc. A QR code is a 2-dimensional barcode localised thr...
详细信息
ISBN:
(纸本)9781509016488
QR codes have become useful and efficient data storage tools which are exploited in many commercial applications including product tracking, website redirection, etc. A QR code is a 2-dimensional barcode localised through three finder patterns (three squares characterised by a series of alternative black and white modules at ratios 1: 1: 3: 1: 1) placed in its three corners. QR codes are generally placed in different environments with complex backgrounds (overlapping text, pictures, etc.), and are often captured under unfavourable conditions such as poor lighting. These factors can significantly affect the recognition ability and thus may hinder correct QR code localisation and identification. In order to appropriately address these issues, in this paper, we present a QR code recognition algorithm based on histogram of oriented gradients (HOG) features combined with support vector machine (SVM) classifiers. Using HOG, we extract gradient features of each extracted pattern. Subsequently, the obtained features are passed to two linear SVM classifiers, one trained with finder patterns and one trained with alignment patterns, to remove irrelevant patterns. QR codes are then conveniently localised according to a pattern closeness constraint. In the laststage, the captured code is enhanced by applying a perspective correction followed by image binarisation and morphological processing. Finally, the patterns are decoded using an accurate 2-d barcode decoder. Our proposed approach is designed for an embedded systems using a Raspberry Pi equipped with a HD camera and a small robot carrying the equipment.
Facial expressions are the most effective way of communication between humans. Hence, recognition of facial expressions is an emerging research topic in the area of imageprocessing and patternrecognition. The task o...
详细信息
ISBN:
(纸本)9789811034336;9789811034329
Facial expressions are the most effective way of communication between humans. Hence, recognition of facial expressions is an emerging research topic in the area of imageprocessing and patternrecognition. The task of recognizing facial expressions is challenging because of variation in many parameters like illumination, pose, ethnicity etc. This article presents an effective approach for recognition of facial expressions with variation in illumination. ROI based Local Binary pattern is used for extracting feature information and Neural Network as classifier. Japanese Female Facial Expressions (JAFFE) database is used for obtaining the results.
processing of the images received from moving objects which including unmanned aerial vehicles (UAVs) was considered. To increase the speed and ensuring high-quality filtering without additional distortions while proc...
详细信息
ISBN:
(纸本)9781509037360
processing of the images received from moving objects which including unmanned aerial vehicles (UAVs) was considered. To increase the speed and ensuring high-quality filtering without additional distortions while processing the double median filter was designed. Interaction between VHDL design environment and FPGA tools during double median filter development was shown. Models of ensuring quality of definition of geometrical features after a filtration during recognition of routes on the image at optical navigation are presented.
Support Vector machine (SVM) is one of the fastest growing methods of machinelearning due to its good generalization ability and good convergence performance;it has been successfully applied in various fields, such a...
详细信息
ISBN:
(纸本)9783662479261;9783662479254
Support Vector machine (SVM) is one of the fastest growing methods of machinelearning due to its good generalization ability and good convergence performance;it has been successfully applied in various fields, such as text classification, statistics, patternrecognition, and imageprocessing. However, for real-time data collection systems, the traditional SVM methods could not perform well. In particular, they cannot well cope with the increasing new samples. In this paper, we give a survey on online SVM. Firstly, the description of SVM is introduced, then the brief summary of online SVM is given, and finally the research and development of online SVM are presented.
作者：
Azmedroub, BoussadOuarzeddine, Mounira
Department of Telecommunication Laboratory of Image Processing and Radiation Bab-Ezzouar Algiers16111 Algeria
Polarimetric Synthetic Aperture Radar (PolSAR) imagery classification is widely investigated, and there are a lot of proposed classifiers. The main issue in the classification of PolSAR images is the extraction of eff...
详细信息
暂无评论