Commodity identification method and device, electronic equipment and readable medium

Abstract

The invention provides a commodity identification method and device, electronic equipment and a readable medium. The method comprises the steps of obtaining two-dimensional feature information and three-dimensional point cloud information of a commodity; when judging that the two-dimensional feature information meets a first condition, identifying the commodity by using a two-dimensional matchingalgorithm; when judging that the two-dimensional feature information and the three-dimensional point cloud information meet a second condition, identifying the commodity by using a three-dimensional matching algorithm; and when judging that the two-dimensional feature information and the three-dimensional point cloud information meet a third condition, identifying the commodity by using a frame matching algorithm. According to the commodity recognition method and device, the electronic equipment and the readable medium, the recognition algorithm can be adaptively matched, and the commodity recognition accuracy is improved.

Description

technical field [0001] The present disclosure relates to the technical field of warehousing and logistics, and in particular to a commodity identification method, device, electronic equipment, and computer-readable medium. Background technique [0002] In the increasingly mature environment of online shopping, e-commerce is required to be equipped with a good distribution process. In the distribution process, it is necessary to automatically distribute the goods according to the customer's order to pack the order. The automatic distribution process is usually realized by automatically identifying and grabbing the goods by the machine. The automatic picking process of the machine can be, for example, the picking process, such as figure 1 As shown, the robot picking in the box refers to the scene where the robot takes out the corresponding quantity of goods from the turnover box according to the requirements according to the picking task issued by the system on the basis of v...

AI Technical Summary

Problems solved by technology

For densely placed products, it cannot effectively identify the gaps between objects, so that the matching accuracy is reduced

In addition, most of the existing products use square packaging boxes, such as 3C products, and are packaged in plastic paper, resulting in weak texture and high reflectivity, which brings great difficulty to accurately obtain features; and the existing The feature matching algorithm is relatively simple, and it cannot intelligently switch to a matching algorithm suitable for the current scene according to the characteristics of the recognition object, resulting in low product recognition

[0004] In the existing technology, there is also a method of using deep learning algorithms to identify products. However, this method requires a large amount of labeled data to train the network, and the cycle of labeling data and training the network is long, which leads to rapid product change. not applicable

Images