Target Detection and Tracking System Based on Embedded PIC32 Microcontroller

By drawing on the successful experience of embedded systems in the fields of electronic technology, signal processing, and computers, and based on the analysis of target detection and tracking algorithms, the target detection and tracking algorithms are combined with embedded technology to design a target detection and tracking system based on the embedded PIC32 microcontroller. The system can realize the miniaturization and intelligence of the target detection and tracking system, and the target identification and tracking are carried out using a specific flying target as an example.

Research purposes

With the continuous expansion and improvement of the demand for military and civilian equipment in modern society, the recognition and tracking technology of moving targets has rapidly developed into a very important technology in the field of modern information processing. It is also the focus and difficulty of UAV field battlefield reconnaissance technology, and plays an irreplaceable role in many fields. At present, the target detection and tracking technology based on PC has become mature, but its application research on embedded platforms is still in its early stages. Since embedded systems have a series of advantages that cannot be replaced by PCs, such as small size and easy to carry, how to effectively detect and track flying targets and implement them in embedded systems is an urgent issue to be solved. In response to the above problems, this paper designs an embedded-based target detection and tracking system. This system has the characteristics of small size, good real-time performance, and can effectively track moving targets in real time. It also makes a useful attempt to carry out the research of real-time and miniaturized embedded machine vision systems.

System overall plan

This topic studies the embedded implementation of UAV target detection and tracking in complex background environments. The target tracking system uses a mobile image processing device to detect and track moving objects in the image, solving the miniaturization requirements of the target tracking system for military and industrial development.

1. The main research contents are as follows:

(1) Hardware design of embedded visual tracking system

Embedded platform is an effective way to solve the real-time and miniaturization problems. The target tracking system solution based on PIC32 embedded platform can meet the functional requirements of the equipment.

In response to the actual needs of miniaturization, this system uses a USB camera as the device for acquiring target images, a PIC32 core development board as the central controller, and peripheral devices such as the pan-tilt head and pan-tilt controller to form a complete visual feedback tracking system.

(2) Software design of embedded visual tracking system

According to the characteristics of the tracked target, a real-time recognition and tracking algorithm is designed. The embedded system software MPLAB IDE integrated development environment is used to develop the system's data reception, feature recognition, feedback control and other software modules. The MPLAB C32 C compiler is used to compile and implement the target tracking algorithm to achieve recognition and tracking of moving targets.

(3) Simulation experiment and debugging

Tracking experiments are carried out on the designed embedded experimental platform, and after repeated debugging, the test tracking effect meets the expected requirements.

2 Study plan 

According to the above research content, the embedded implementation of target detection and tracking solution is divided into two parts: hardware and software.

Hardware solution for embedded visual tracking system:

The system hardware structure diagram is shown in Figure 1:

Figure 1 System hardware structure diagram

The overall hardware structure of the system is shown in Figure 1. The MPLAB IDE integrated development environment is ported and configured on the PIC32 embedded system platform. According to the characteristics of the platform and application, a suitable file system is created and an application program is written. The moving target is captured in real time by the USB camera and sent to the PIC32 processor. The powerful computing power of the embedded processor is used to process the captured image, complete target recognition and positioning, and control the pan-tilt controller to adjust the camera posture so that the camera is aimed at the moving target to achieve real-time tracking. Good real-time performance and compact size are the goals pursued by this embedded tracking system.

Software solutions for embedded visual tracking systems:

(1) MPLAB C32 C compiler

C code application: The 32-bit language tool libraries are located in the pIC32mxlib subdirectory of the MPLAB C32 C compiler installation directory, which is stored by default in: C:Program FilesMicrochipMPLAB C32pic32mxlib. These libraries can be linked directly into the application through the MPLAB C32 linker.

(2) Startup code

To initialize the variables in data memory, the linker creates a data initialization image. This image must be copied into RAM at startup, before the application properly gains control. The startup code in crt0.o performs the initialization of the runtime environment.

(3) 32-bit peripheral function library

The 32-bit software and hardware peripheral libraries provide functions and macros for setting up and controlling 32-bit peripherals. These libraries are processor-specific and are in the form of libmchp_peripheral_Device.a, where Device is the 32-bit device model.

(4) Standard C function library (including mathematical functions)

A complete set of ANSI-89 compliant libraries is provided. The standard C language library files are libc.a (written by MIPS Technologies), libe.a, and libm.a. A typical C application must include all three libraries, which are linked in by default without user specification.