论文部分内容阅读
针对数字信号处理(DSP)技术相关课程实践应用性很强的特点,结合TI公司DaVinci系列DSP芯片的数字视频监控等最新应用,开发了基于SEED-DTK6437硬件平台的运动目标识别与跟踪实验系统,分别采用二帧差法和改进型三帧差法实现了运动目标识别,并利用形心算法实现了运动目标的跟踪。同时,将该系统有效地应用到宁波大学第三学年短学期课程设计和虚拟企业校内集中毕业实习等实践教学环节,为培养学生DSP芯片软硬件系统设计能力提供了良好的实验方案。
Aiming at the characteristics of practical application of digital signal processing (DSP) related courses and the latest application of digital video surveillance such as TI’s DaVinci series DSP chip, a moving target recognition and tracking system based on SEED-DTK6437 hardware platform is developed. The two-frame difference method and the improved three-frame difference method are respectively used to realize the moving object recognition. The moving object tracking is realized by the centroid algorithm. At the same time, the system is effectively applied to the practical teaching such as the short semester course design of the third year of Ningbo University and the intensive graduation practice in the virtual enterprise, which provides a good experimental program for training students’ DSP hardware and software system design capabilities.