Research And Development Of Car Door Lock Control System Based On Palmprint Recognition And LIN Bus

With the fast enhancement of car parc in China,the study of car anti-theft is attached more and more importance now,according to the rapid development of the national economy and the continuous improvement of people’s living standards.In order to improve the anti-theft performance of vehicles,this paper presents a new control system for automobile door locks,which combines the palmprint recognition technology with the LIN bus network technology to control the automobile door locks.In this system,the LIN network data communication is implemented by the master node controller together with the slave node controller to realize the control of the automobile door lock system.The palmprint recognition module is designed to perform data matching of palmprint features.The matching result is received by the slave node controller and converted to LIN bus frame format,which is then sent to the bus and received by the master node controller.After that,the master node controller analyzes the frame and sends commands to execute the control of the door lock.This system effectively improves the anti-theft performance,security and flexibility of automobiles and exhibits practical value and economic value in the prevention of auto thefts.Firstly,this paper studies the palmprint recognition algorithm.Aiming at the problem that the illumination,palm position,acquisition equipment and other factors of the system will affect the recognition rate of palmprint,and the problem of high calculation complexity when using the traditional sparse reconstruction method,a novel palmprint recognition method is put forward based on bidirectional 2D principal component analysis((2D)~2PCA)and compression sensing.This method uses(2D)~2PCA to reduce the dimension of the palmprint image in both the rows and the columns,and extract the feature matrix as the over-complete dictionary for the compression sensing algorithm.In order to solve the sparse representation of the over-complete dictionary,this paper proposes an improved orthogonal matching pursuit algorithm(OMP)to obtain the sparse reconstructed image,and to solve the minimum residual error between the test image and the reconstructed image,which is then compared to complete the recognition process.The experimental results show that this method can effectively reduce the complexity and narrow the calculation time of palmprint recognition.A high palmprint recognition rate can be obtained even when the light and palm position change.Secondly,this paper produces a preliminary design of the system implementation,including the system hardware design and software design,and the system has been tested.The hardware design includes the system LIN node design,system processor selection,storage part design,image acquisition design and system’s local circuit design.The software design includes the software workflow design of system LIN bus network module and the software design of the palmprint recognition module.In the final test of the system,the hardware debugging and communication testing of the system palmprint recognition module have verified the feasibility of the system.Finally,this paper summarizes the research and puts forward prospects for the following work.