System and method for image classification

A method of offline image classification in a mobile device with storage capacity of 10-50 megabytes, the method comprising:

• linking between a plurality of predetermined objects and categories, such that each object corresponds to at least one category;
• assigning, based on an output of a categorization algorithm, a set of attributes for each object until hierarchical relationships between all categories are determined, wherein at least one predetermined attribute of each object relates to at least one category;
• storing at least one classified image for each object in an object database on the mobile device, wherein each classified image corresponds to at least one predetermined attribute;
• capturing an unclassified image, by an imager of the mobile device;
• comparing, by a processor of the mobile device, the captured unclassified image to the at least one classified image, wherein the comparing is carried out offline;
• determining, by the processor, at least one attribute of the captured unclassified image, based on the comparison and based on the assigned attributes for each object, wherein the determining is carried out offline;
• assigning, by the processor, a classification to the captured unclassified image, based on:
• the output of the categorization algorithm, the at least one determined attribute, and the determined hierarchical relationships between the categories, wherein the assigning is carried out offline;
• and
• displaying the assigned classification,
• wherein the categorization algorithm, when executed by the processor, causes the processor to process captured unclassified images and classify the images to categories, the classification algorithm having been trained on classified images and objects from the object database.