Abstract: The invention provides an English word image recognition method, mainly loading a to-be-recognized image and performing a one-dimensional convolutional neural network operation and a fully connected operation processing to generate a feature map, outputting the feature map by a bidirectional long short-term memory (LSTM) network and performing a fully connected operation to generate a feature map, then performing a probability recognition and outputting a probabilistic string, and then recognizing the probabilistic string and outputting a word recognition result to solve the problem of producing a large amount of operation in the conventional two-dimensional recognition operation, thereby achieving efficacies of reducing costs of recognition equipment and enabling fast and accurate recognition.

Abstract: The invention provides an English char image recognition method mainly generating a rectangular coordinate frame from a loaded English character image, finding a gravity center in the rectangular coordinate frame, obtaining feature points on the rectangular coordinate frame and around the gravity center, performing one-dimensional convolutional operation and processing on the feature points to generate six layer feature maps, and generating a character recognition result from the six layer feature maps to solve the problem of generating a large amount of computation in conventional two-dimensional recognition operation, thereby achieving an efficacy of reducing recognition equipment costs and enabling fast and accurate recognition.

Abstract: The invention provides an E13B image interpretation method mainly aiming at interpreting numbers and symbols in a check reading code printed with magnetic ink on a check, and defining a plurality of feature areas for each of the numbers and dividing each of the numbers into a plurality of blocks, and then interpreting according to whether an amount of magnetic ink being in the feature areas and a magnetic ink density between the blocks, and also generating a plurality of target pixels for each of the symbols, and then interpreting according to a relative position and a size of the target pixels.

Abstract: The invention provides an English char image recognition method mainly generating a rectangular coordinate frame from a loaded English character image, finding a gravity center in the rectangular coordinate frame, obtaining feature points on the rectangular coordinate frame and around the gravity center, performing one-dimensional convolutional operation and processing on the feature points to generate six layer feature maps, and generating a character recognition result from the six layer feature maps to solve the problem of generating a large amount of computation in conventional two-dimensional recognition operation, thereby achieving an efficacy of reducing recognition equipment costs and enabling fast and accurate recognition.

Abstract: The invention provides an E13B image interpretation method mainly aiming at interpreting numbers and symbols in a check reading code printed with magnetic ink on a check, and defining a plurality of feature areas for each of the numbers and dividing each of the numbers into a plurality of blocks, and then interpreting according to whether an amount of magnetic ink being in the feature areas and a magnetic ink density between the blocks, and also generating a plurality of target pixels for each of the symbols, and then interpreting according to a relative position and a size of the target pixels.