Character recognition (CR) considered as one of the most important in the field of patternrecognition. the ultimate objectives of the Optical Character recognition (OCR) system is to simulate the capability of readin...
详细信息
ISBN:
(纸本)9781538639603
Character recognition (CR) considered as one of the most important in the field of patternrecognition. the ultimate objectives of the Optical Character recognition (OCR) system is to simulate the capability of reading, hence the OCR considered as artificial intelligence. In this paper, a character-handwritten recognition for the Arabic language is developed. the main aim of the system is to save time and effort Arabic OCR. In addition, to be the alternative of the typing manual due to provide it fast and reliable. the system has four main stages;preprocessing, segmentation, feature extraction, classification, and recognition. the system is off-line and depends on the image acquisition. So, after acquitted the image has to go through the main stages. the Neural Network used as a classifier. the proposed system is able to recognize as many characters as can with high accuracy rate. In addition, it is focusing on the character that has similarities and the system will also be considered about the number of dots and its position, and the connected components.
the olfactory system is a remarkable system capable of discriminating very similar odorant mixtures. this is in part achieved via spatio-temporal activity patterns generated in mitral cells, the principal cells of the...
详细信息
ISBN:
(纸本)9783642213267
the olfactory system is a remarkable system capable of discriminating very similar odorant mixtures. this is in part achieved via spatio-temporal activity patterns generated in mitral cells, the principal cells of the olfactory bulb, during odor presentation. In this work, we present a spiking neural network model of the olfactory bulb and evaluate its performance as a patternrecognition system with datasets taken from both artificial and real pattern databases. Our results show that the dynamic activity patterns produced in the mitral cells of the olfactory bulb model by pattern attributes presented to it have a pattern separation capability. this capability can be explored in the construction of high-performance patternrecognition systems.
In this study, CiteSpace visualization analysis tool is used to analyze relevant literatures on core journals of patternrecognition algorithms in CNKI database. A total of 511 papers from Chinese core journals in the...
详细信息
A method for 3-D measurement using a stripe line pattern has been developed and tested. Both fundamentals and experiment result are presented in the paper. Only one projected pattern is used during measurement and onl...
详细信息
A general purpose software package "PREDITAS" is presented, aimed at solving a wide range of patternrecognition and diagnostic problems with respect to constraints and requirements imposed by practice. the ...
详细信息
A neural patternrecognition system using tap tone signals has been developed to diagnose cracks in coil insulation. the tap tone signals are converted from a wave shape into a frequency spectrum by a fast Fourier tra...
详细信息
A neural patternrecognition system using tap tone signals has been developed to diagnose cracks in coil insulation. the tap tone signals are converted from a wave shape into a frequency spectrum by a fast Fourier transform analyzer, then discriminated by a computer in which a neural patternrecognition software is installed. Results from a 1/3 octave frequency spectra of tap tone signals agree with electrical characteristics, and show that the system is effective in evaluating cracks in the coil insulation layer.
Local gene network pattern (LGNP) is a useful tool to display complex relationship between multiple genes. A diabetesspecific local gene network (LGN) was constructed with 29 genes (including 9 diabetes susceptibility...
详细信息
ISBN:
(纸本)9781424447138
Local gene network pattern (LGNP) is a useful tool to display complex relationship between multiple genes. A diabetesspecific local gene network (LGN) was constructed with 29 genes (including 9 diabetes susceptibility genes and 20 extra genes to join them) that participate in 30 biological pathways. Further analysis indicated diabetes-specific LGN has close relations with carbohydrate metabolism, lipid metabolism, amino acid metabolism, cofactor and vitamin metabolism. Especially, molecular links demonstrated diabetes was associated with alanine and nicotinamide metabolism, consistent with previous reports. Preliminary six microarray data revealed that some key nutrient metabolism related genes were de-regulated in insulin treated muscles, providing clues how to transform diabetesspecific-LGNP with personalized nutrient formula in future.
1H NMR-based metabonomic method was used to investigate the biochemical variability in serum from rats treated by nephrotoxins. the toxic effect of the model toxins, sodium chromate (Na2CrO4, 20 mg/kg body mass), 2-br...
详细信息
Brain dynamics were analyzed during decision making using human electroencephalographic signals. We sought to identify the pattern of brain activity for actions with and without decision-making, while subjects engaged...
详细信息
ISBN:
(纸本)9781424447138
Brain dynamics were analyzed during decision making using human electroencephalographic signals. We sought to identify the pattern of brain activity for actions with and without decision-making, while subjects engaged in an instrumental reward based learning task Event related potentials (ERPs) were analyzed for reference trials (no choice required) and decision trials. To detect brain activity during decision making, classification was applied to classify reference and decision trials. Support vector machine (SVM) with a nonlinear kernel function was used as a classifier. Classification performance was analyzed across subjects and channels to identify brain regions underlying decision-making. For most subjects, we found that reference and decision trials could be classified with greater than 85% accuracy. ERPs from frontocentral areas of the scalp provided, in general, best classification rates. thus ERPs and SVM classifiers can be used to non-invasively detect decision making in humans.
We studied changes in landscape patterns, especially the trend of dessication at Zhalong Wetland using remote sensing, GIS techniques and analysis of a time series of satellite images from 1979-2006. Our results show:...
详细信息
ISBN:
(纸本)9781424447138
We studied changes in landscape patterns, especially the trend of dessication at Zhalong Wetland using remote sensing, GIS techniques and analysis of a time series of satellite images from 1979-2006. Our results show: (1) the landscape pattern at Zhalong changed significantly. (2) the centroid of wetland shifted from northeast to southwest from 1979-2006, indicating the distribution of water changed. (3) the wetland is getting drier, especially from 1999 to 2006. About 100 km(2) changed from open water to other wetland types of land cover and 611 km(2) marshes changed to drier types of land cover. the wetland has become more fragmented. the area covered by alkalized soil has been expanding. the total alkaline soil area increased by 28.08 km(2), including increase in the core area. Zhalong is a globally important wetland supporting the largest breeding population of Red-crowned Cranes and many other birds. this study of landscape changes is now helping to guide wetland restoration at Zhalong. Reserve management needs to address water supply and wetland fragmentation.
暂无评论