METHODS OF PREPROCESSING OF DIGITIZED HANDWRITTEN DOCUMENTS
Annotation
The problem of automating the analysis of handwritten documents is solved. It is shown that artificial neural networks capable of recognizing images after training on the original data set are used to solve such problems. At the same time, the quality of recognizing new images largely depends on the stage of pre-processing of digitized handwritten documents. A particular preprocessing problem is considered - removing cell lines from an image of a notebook sheet. Four methods of image filtering are analyzed using the OpenCV library of the Python language. A neural network of convolutional architecture is trained to recognize handwritten characters. The work of the trained neural network on documents preprocessed by different algorithms is demonstrated.
Keywords
Постоянный URL
Articles in current issue
- ANALYTICAL APPROACH TO THE SELECTIVE SEARCH OF STATE PROBABILITY FUNCTIONS FOR MARKOV PROCESSES
- ADJUSTING THE PARAMETERS OF THE ADAPTIVE REGULATOR IN SUPER TWISTING SLIDING MODE CONTROL FOR A CLASS OF UNCERTAIN NONLINEAR SYSTEMS BASED ON JAYA ALGORITHM
- METHODS OF PREPROCESSING OF DIGITIZED HANDWRITTEN DOCUMENTS
- REDUCING HARDWARE COSTS OF A LOCAL FRAGMENTED CONTROL DEVICE IN A NEURAL NETWORK ANALOG-TO-DIGITAL CONVERTER
- APPLICATION OF SACRIFICIAL LAYERS IN THE MANUFACTURE OF MULTILAYER Al–Ni FOIL WITH THE EFFECT OF SELF-EXPANDING HIGH-TEMPERATURE SYNTHESIS
- USING A MULTI-BAND DIFFERENTIAL ABSORPTION LIDAR FOR EARLY DETECTION OF FOREST FIRES
- MACRO- AND MICROSTRUCTURAL ANALYSIS OF STRUCTURAL BEARING STEEL SHKH15SG IN THE CONTEXT OF PREDICTIVE DIAGNOSTICS
- SCALING THE PRACTICE OF USING CAD/CAE/CAM SYSTEMS IN SMALL ENTERPRISES
- DEPENDENCE OF SURFACE ROUGHNESS OF TUNGSTEN CARBIDE ALLOYS ON LASER ABLATION PARAMETERS
- THE USE OF MATHEMATICAL MODELS OF OVERHEAD POWER TRANSMISSION LINE ELEMENTS FOR EQUIPMENT DIAGNOSTICS