Rapid commodity settlement system based on RGBD information and deep learning

Abstract

The invention relates to the technical field of machine learning, computer vision, rapid commodity settlement and the like, in particular to a rapid commodity settlement system based on RGBD information and deep learning. The rapid commodity settlement system consists of a hardware part and a software part, wherein the hardware part consists of a conveyor belt, a commodity, a settlement gate, a camera and an infrared sensing device; the software part uses PC and other devices which provide a program operating environment as main carriers; the commodity is placed on the conveyor belt; the conveyor belt is a commodity conveying device; the settlement gate is arranged at one end of the conveyor belt; the infrared sensing device is arranged on the inner wall of the settlement gate; the infrared sensing device is connected with the settlement gate in an inlaid way. Through structural improvement, the rapid commodity settlement system is low in requirement on a hardware environment, low in cost, high in commodity settlement efficiency, free of an unrecognizable condition caused by information label damage, high in stability, high in recognition accuracy and good in recognition effect, sothat problems and defects in the existing technology are effectively solved.

Description

technical field [0001] The present invention relates to technical fields such as machine learning, computer vision, and commodity fast settlement, and in particular to a commodity fast settlement system based on RGBD information and deep learning. Background technique [0002] Image stitching technology is the technology of stitching several overlapping images (which may be obtained at different times, different viewing angles or different sensors) into a large seamless high-resolution image. Many times, due to the limited resolution of ordinary cameras, capturing images with a larger field of view will seriously distort the target area. In order to obtain images with larger viewing angles without reducing the image resolution, the technology of image mosaic using computer has become the research focus of computer graphics, and is widely used in space detection, remote sensing image processing, medical image analysis, video compression, etc. and transmission, virtual realit...

AI Technical Summary

Problems solved by technology

[0042] The purpose of the present invention is to provide a fast commodity settlement system based on RGBD information and deep learning to solve the barcode technology proposed in the background technology: only one commodity can be scanned and identified; it is easily damaged and cannot be recognized normally; the barcode Only limited product information is included in the product, and it needs to be actively written; the field of view is limited, and the barcode needs to be aligned to the scanning area to complete the identification. RFID technology: RFID tags are expensive; product information must be written actively; RFID tags are easy When it is applied to metal and liquid products, the recognition effect may be affected. "Amazon GO": the whole system is very expensive to build, and the store needs to be remodeled; the requirements for algorithm capabilities are very high, and technical It is difficult to implement; when there are too many people in the store, the amount of calculation will be large, and problems and deficiencies of failure are prone to occur

Images