Integrated license plate face recognition method and module based on Raspberry Pi

Abstract

The invention relates to an integrated license plate face recognition method and module based on Raspberry Pi, and the method is characterized in that the method comprises the following steps: carrying out the environment building on the Raspberry Pi; extracting a license plate image and a face image for training, and transplanting license plate recognition and face recognition to the Raspberry Pi; binding the license plate recognition service logic and the face recognition service logic; extracting a to-be-identified face or license plate image for identification; when the face or license plate image to be identified is judged to be the same face or license plate, the Raspberry Pi sends an opening instruction to a controller of an access control or a barrier gate through the server; according to the invention, license plate recognition and face recognition are integrated on the basis of the Raspberry Pi; the module is simple in structure, small in size, low in cost and most suitable for places where old communities, face access control and license plate gates are very close to one another, the gate and the access control are connected through the access control device, the licenseplate recognition control over the gates and the face recognition control over the gates are achieved, the use mode is convenient, and the recognition rate is high.

Description

technical field [0001] The present invention relates to the field of intelligent management systems based on license plate recognition and face recognition, in particular to an integrated license plate face recognition method and module based on Raspberry Pi. Background technique [0002] License plate recognition and face recognition have been widely used in the industry and there are many successful use scenarios. Most of the current face recognition and license plate recognition are algorithms based on traditional algorithms plus machine learning. This algorithm requires The amount of calculation is small, but the accuracy is poor, which is stretched in many fields and cannot meet the needs of customers. [0003] Although the recognition algorithm based on deep learning is much higher than the original algorithm in terms of recognition accuracy, such algorithms often require strong computing power, so generally a server is required for unified calculation, or the algorith...

AI Technical Summary

Problems solved by technology

[0002] License plate recognition and face recognition have been widely used in the industry and there are many successful use scenarios. Most of the current face recognition and license plate recognition are algorithms based on traditional algorithms plus machine learning. This algorithm requires The amount of calculation is small, but the accuracy is poor, and it is stretched in many fields and cannot meet customer needs

[0003] Although the recognition algorithm based on deep learning is much higher than the original algorithm in terms of recognition accuracy, such algorithms often require strong computing power, so generally a server is required for unified calculation, or the algorithm is transplanted to the FPGA board In fact, the cost of these computing boards is very high due to the small number of developers and small demand. In addition, in order to recover the development cost and protect the system, the manufacturers have adopted the integration of the camera and the FPGA board for sale. Such measures make the product price further At the same time, in terms of program use and adjustment, the scalability and application scenarios are further reduced. In the end, high prices and low application scenarios are difficult to meet the increasingly demanding customer needs.

In addition, in general scenarios, the products on the market are separated from the license plate and the face. If the user needs two functions, they need to purchase two sets of independent products separately, and the business logic itself cannot effectively cooperate to form a single product. system, which will undoubtedly further increase the user cost

Images