| In the military training, shooting is a basic task, shooting level directly affects the fighting capacity of the army. But, until now, the shooting training still rely on artificial indication. Human consumption, low efficiency and poor accuracy all seriously influence the improvement of soldiers’shooting levelBut with the improvement of science and technology, now there are some automatic target scoring system, like audio-electronic target reporting system, optical-electronic automatic target reporting system. But its high cost and inconvenience make them difficult to spread. In recent years, Automatic target-scoring system based on image processing get more and more attention. Because the structure of hardware is simple and it mainly relay on the software. So it can be updated easily. But now the system design is not reliable. It has poor adaptability, low recognition rate, so it is not widely used.To solve the above problems, this paper designed an automatic target system based on DSP image processing. In terms of hardware, using TI’s TMS320DM6437as the core of image processing platform, which can realize the following functions:image capturing, image processing, data transmission and displaying the results. In terms of software, designed a new bomb detection methods, including relying information of color and gray together to binary image, carried out mathematical morphology operation for the details, and get the target circles by Canny edge detection algorithm, etc. These methods can improve the performance for different quality images, improve accuracy of the system and make the system more practical. Meanwhile, the system can form a training system when interconnected with multiple nodes through the CAN bus. Then it has the ability of sending up bullet information, remote monitoring, scalability.The system has high precision, high speed, small size, scalability, and is easy to upgrade, etc. It can meet the requirements of military training well. |
PDF Full Text Request