Tracing method and system based on biological image feature recognition

Abstract

The invention discloses a tracing method and system based on biological image feature recognition. The tracing method comprises the steps: training the feature extraction models of a biological product in various forms; establishing a tracing information database, and storing tracing information of the biological product in the respective forms; acquiring the tracing information: acquiring a firstbiological product picture and product flow information, extracting a first biological image feature from the first biological product picture, forming tracing information, and storing the tracing information in a tracing information database; searching for the tracing information: acquiring a second biological product picture, extracting a second biological image feature from the second biological product picture, and retrieving the tracing information database by taking the second biological image feature as an index to obtain the tracing information of the biological product matched with the second biological product picture; and associating the tracing information : associating the tracing information of the previous and following forms of the biological product. According to the method, the cost of the existing tracing technology can be reduced, the confidentiality is better than that of the entity label, and the problem that the entity label may be lost is solved.

Description

technical field [0001] The invention relates to the technical fields of computer vision, deep learning target recognition and traceability system, in particular to a traceability method and system for identifying biological image features using deep learning. Background technique [0002] With the development of science and technology, the traceability technology of biological products is becoming more and more mature, and the traceability system is becoming more and more perfect. This technology is of great significance to food safety, self-regulation of producers and tracking of product information by consumers. At present, when the world is affected by the new crown epidemic, the traceability of fresh and living organisms has greatly helped the development of epidemic control. After the outbreak, it can also help track and locate the source of the epidemic to a certain extent. At the same time, traceability of biological products can also reduce unnecessary losses. Afte...

AI Technical Summary

Problems solved by technology

[0004] 1. Entity labeling inevitably has a certain cost

[0005] 2. Falsification of physical labels exists, and unscrupulous traders seek profit to resell labels of products with higher economic value. At the same time, in order to prevent this situation, manufacturers need to continuously update labeling technology, which will cause greater costs

[0006] 3. The physical label may be lost during transportation. After the loss, manpower and material resources are required to re-encode the unique identity and complete the traceability information

Images