Recognition of handwritten Marathi character/digits is comparatively a tough task as compared to English. It has several types of applications including the postal code reading and sorting, banks check amount processing. In this paper a novel feature extraction and selection method is proposed for the recognition of isolated handwritten Marathi numbers which is based on one dimensional Discrete Cosine Transform (DCT) algorithm. The scanned document is pre-processed and segmented to create isolated numerals. Features for each numeral can be calculated after normalizing the numeral image to 32 × 32 size. Based on these reduced features, the numerals are classified into appropriate groups. Neural network is used for classification of numerals based on the extracted features. The results of proposed method are compared with the results obtained using Discrete Wavelet Transform and zonal discrete cosine transform (ZDCT). The proposed approach gives improved results as compared to zonal DCT and DWT method. Experimental results shows accuracy observed for the method is 90.30% with normalized numeral image of size 32 × 32.
Working at Bell Labs in 1950, irritated with error-prone punched card readers, R W Hamming began working on error-correcting codes, which became the most used error-detecting and correcting approach in the field of channel coding in the future. Using this parity-based coding, two-bit error detection and one-bit error correction was achievable. Channel coding was expanded further to correct burst errors in data. Depending upon the use of the number of data bits ‘d’ and parity bits ‘k’ the code is specified as (n, k) code, here ‘n’ is the total length of the code (d+k). It means that 'k' parity bits are required to protect 'd' data bits, which also means that parity bits are redundant if the code word contains no errors. Due to the framed relationship between data bits and parity bits of the valid codewords, the parity bits can be easily computed, and hence the information represented by 'n' bits can be represented by 'd' bits. By removing these unnecessary bits, it is possible to produce the optimal (i.e., shortest length) representation of the image data. This work proposes a digital image compression technique based on Hamming codes. Lossless and near-lossless compression depending upon need can be achieved using several code specifications as mentioned here. The achieved compression ratio, computational cost, and time complexity of the suggested approach with various specifications are evaluated and compared, along with the quality of decompressed images.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.