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.
This paper presents a character recognition system of handwritten Odia script. Although, a lot of research has been carried out in handwritten character recognition, not considerable research has been carried out to recognize handwritten Odia characters. In this work we have attempted to classify 36 basic Odia consonants. Total 48 geometric features were extracted which includes shadow features, centroid...
Deep convolutional neural networks (DCNN) have recently achieved state-of-the-art performance on handwritten Chinese character recognition (HCCR). However, most of DCNN models employ the softmax activation function and minimize cross-entropy loss, which may loss some inter-class information. To cope with this problem, we demonstrate a small but consistent advantage of using both classification and...
Sign language is widely used by individuals with hearing impairment to communicate with each other conveniently using hand gestures. However, non-sign-language speakers find it very difficult to communicate with those with speech or hearing impairment since it interpreters are not readily available at all times. Many countries have their own sign language, such as American Sign Language (ASL) which...
the gray gradient co-occurrence matrix based Uyghur document image retrieval method is proposed in this paper. 15 parameters of gray gradient co-occurrence matrix such as small gradient advantage, high gradient strengths, intensity in homogeneity etc are extracted separately, and document image features are matched with the characteristic distance classifier and the Euclidean distance classifier....
The script ‘Devanagari’ is used in many Indian languages. Hindi language is also under Devanagari script. In this paper recognition of Hindi characters is done by using a three step procedure. First step is preprocessing, in which binarization of the image and separations of characters are performed. Each Hindi word has a horizontal bar on the top of word. That bar is also removed in preprocessing...
Present a novel method of character stroke feature extraction based on the histogram of gradient angles. Firstly, produce the gradient angle matrix of the pixels in the character edge by sifting out the gray image using the SOBEL operator. Secondly, set up the character stroke features as sub-vectors of each block matrix using the relationship between the gradient angle matrix and the character strokes...
This paper describes a method for constructing the most efficient and robust coarse classifier from a large number of basic recognizers which are obtained by different parameters of feature extraction, different discriminant methods or functions, and so on. The architecture of the coarse classification is a sequential cascade of basic recognizers and reduces the candidates after each basic recognizer...
In this paper a new approach for Farsi font recognition is presented. In this approach using the Euclidean distance between spatial descriptors and gradient value in each point of boundary of some especial letters, font of a document image is recognized. To test this approach we constructed a dataset consisting of some letters in 20 widely used Farsi fonts and obtained recognition rate more than 94%...
The objective of this project is to build a character recognition system, which is able to recognize printed and handwritten character from A to Z. the typical optical character recognition systems, regardless the character's nature, are based mainly on three stages, preprocessing, features extraction, and discrimination. Each stage has its own problems and effects on the system efficiency which is...
This paper presents a novel approach for recognition of unconstrained handwritten Marathi characters. The recognition is carried out using multistage feature extraction and classification scheme. The initial stages of feature extraction are based upon the structural features and the classification of the characters is done according to their parameters. The final stage of feature extraction employs...
To recognize protuberant characters on the labels, a novel recognition method based on gray image using wavelet packet transform (WPT) and improved Singular Value Decomposition (SVD) is suggested. The images are decomposed by WPT, and then the singular values (SVs) feature and projection feature of approximation images obtained through the above process are combined to form a new eigenvector using...
Arabic text recognition offers unique technical challenges and has been addressed more recently in the document analysis research field than other languages. Automatic Arabic/Indian numeral Optical Character Recognition (OCR) system for postal services are used in many countries, but still there are problems in such systems where machines still provide errors in reading the crucial information needed...
In recent years research towards Indian handwritten character recognition is getting increasing attention. Many approaches have been proposed by the researchers towards handwritten Indian character recognition and many recognition systems for isolated handwritten numerals/characters are available in the literature. To get idea of the recognition results of different classifiers and to provide new...
In the Chinese character intelligent formation, the same basic element has different position, size and shape in different Chinese characters. The transformation from the basic element to the Chinese character can be realized by affine transformation, and a novel approach was proposed to acquire the transformation coefficients of basic elements. The interested region was determined based on the structure...
An effective algorithm for number and letter character recognition is proposed in this paper. Our algorithm employs template matching, but it unlike traditional template matching method using the original pixel value to match. Our algorithm draws some features from the original image, and then obtains an eigenvector of 192 dimensions. Before drawing features, the image is disposed using math morphologic...
An approach of vehicle license plate recognition (VLPR) based on Gabor filter is presented in this paper. The parameters of Gabor filter are different from most traditional ways in the field. The novel way enables us to concisely model the statistical dependencies, and achieve a more reliable and local model using blocks. Firstly, the accurate position of license plate must be extracted from the original...
An applied coarse classification scheme for handwritten Chinese character is presented in this paper. Four-side code feature is employed as coarse feature and RBF neural network is used as classifier in this experiment. In contrast to Euclidean distance as the measurement of similarity used in conventional method, RBF (radial basis function) neural network is better to fit the data of each class....
We proposed an on-line writer dependent Khmer recognition method based on FIR system characterizing hand-writing motion. The handwriting motion can be described by two features, barycenter trajectory and its velocity. The barycenter is determined from the center point of the script and the two adjacent pen-point positions with respect to time in handwriting process. Then the barycenter and its velocity...
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.