Information processing device, store system, and information processing method

Abstract

The present invention provides an information processing device, a store system, and an information processing method capable of improving operability. The information processing device includes: a reference value changing unit for reducing the reference value when the color attribute of the frame image including the image of the object captured by the imaging unit does not exceed a predetermined reference value; When the color attribute exceeds a predetermined reference value, calculate the degree of similarity to what extent the image representing the object contained in the frame image is similar to the reference image of each registered product registered in the thesaurus; the product candidate extraction unit, based on the similarity, extracting candidates of registered goods corresponding to the object; and the determination unit, when the process of lowering the reference value is not performed by the reference value changing unit for the candidates of the registered goods, or when the process of lowering the reference value is performed by the reference value changing unit for the candidates of the registered goods In the case of a dark-colored commodity, the candidate for the registered commodity is determined as the registered commodity corresponding to the object.

Description

[0001] This application claims the priority of the Japanese application with the filing date of February 29, 2012 and the application number JP2012-044238, and cites the content of the above application. technical field [0002] The invention relates to an information processing device, a store system and an information processing method. Background technique [0003] At present, there is known a technique related to general object recognition. From the image data of the target item (object) captured by an image sensor device, the feature value of the target item is extracted and compared with the previously prepared lexicon. The verification uses the data (features) to compare and calculate the similarity, and according to the calculated similarity, identify (detect) the category of the item, etc. In addition, a store system is proposed in which the technology related to general object recognition is used for recognition of products such as fruits and vegetables, and sales ...

AI Technical Summary

Problems solved by technology

[0005] However, for example, when a deep-colored object such as an eggplant is used as an object recognition object, since it is difficult to have a difference in color attributes (such as brightness) and the difference from the background frame image is small, there is a possibility that the object cannot be recognized. Circumstances of identification processing

[0006] In addition, even when the object recognition process can be started, it may be misrecognized because the difference with the background frame image is small.

Images