Cargo tray detection system and method based on TOF (Time of Flight) camera

Abstract

The invention provides a cargo tray detection system and method based on a TOF (Time of Flight) camera. The system comprises a central control unit and the TOF camera, wherein the TOF camera is used for acquiring an RGBD image and transmitting the RGBD image to a central processing unit; the central processing unit is used for converting the RGBD image into a gray level depth image and taking a detected regular shape as a tray plane candidate region; if the tray plane candidate region contains at least two rectangular regions, the rectangular regions are used as branch hole candidate regions and pixel area of the branch hole candidate regions and central point coordinates of the branch hole candidate regions are calculated; an area ratio tij between any two branch hole candidate regions Ri and Rj in the tray plane candidate region is calculated; when two Ri and Rj which accord with a threshold value range are screened and remained, a cargo tray is detected; a middle point Oij of a connecting line of the central points of the remained Ri and Rj is used as the central point of the surface of the cargo tray; distance value d between the Oij and the TOF camera is obtained through gray level value of the Oij. The cargo tray detection system and method based on the TOF camera have the advantages of high precision, low price, strong robustness and the like.

Description

technical field [0001] The present invention relates to the technical field of industrial robots, in particular to a cargo pallet detection system and method based on a TOF camera. Background technique [0002] With the development of modern logistics technology, automatic guided vehicles (AGV) are playing an increasingly important role in intelligent warehousing technology, and the detection of cargo pallets is one of the core technologies of AGV. The storage environment has the characteristics of complex background, unstable light conditions, and many dynamic / static obstacles. Accurate and efficient detection and positioning of cargo pallets is an urgent problem to be solved at present. [0003] After patent retrieval, the current detection of cargo pallets mainly adopts visual detection, laser radar detection and other methods. Among them, although the cost of monocular vision is not high, the accuracy is not high and is greatly affected by light; both multi-line lidar a...

AI Technical Summary

Problems solved by technology

Among them, although the cost of monocular vision is not high, the accuracy is not high and is greatly affected by light; both multi-line lidar and binocular vision detection have the characteristics of high precision, but multi-line lidar is too expensive, and double-sided vision detection is affected. The impact of light is large; the cost of single-line lidar is relatively low, but the amount of information obtained is insufficient, and false detections are prone to occur; the visual detection method based on feature tags requires relatively high accuracy for label laying, and poor environmental robustness, etc.

Images