Character recognition method

A method for recognizing characters in an image of a document comprising at least one alphanumeric field, the method comprising the steps consisting in:

• segmenting the image to identify objects therein;
• defining a bounding box around each object and making a first selection to select the bounding boxes supposedly containing a character based on at least one theoretical dimensional characteristic of an alphanumeric character;
• making a second selection comprising applying shape descriptors to each firstly selected bounding box and implementing a decision making algorithm to select, on the basis of the descriptors, the firstly selected bounding boxes supposedly containing a character;
• grouping the secondly selected bounding boxes according to relative positions of the secondary selected bounding boxes;
• making a third selection by dividing each of these secondly selected bounding boxes into a plurality of cells for each of which a texture descriptor in the form of an histogram of oriented gradient is determined, the histograms then being concatenated and a decision making algorithm being implemented to select, on the basis of the descriptors, the secondly selected bounding boxes supposedly containing a character;
• performing a character recognition on the thirdly selected bounding boxes.