An application of video motion detection technology in city traffic management

For city traffic management, an underground induction winding is used to identify vehicles inside the red light forbidden zone in most computer monitoring systems at traffic lights. We described how to apply video motion detection (VMD) technology in these computer surveillance systems instead of underground induction windings. Using VMD eliminates the underground induction windings that damage road surfaces. The possibility of applying and implementing VMD technology with supporting hardware was analyzed. We described how to develop the program to record automatically those vehicles that violate traffic rules at traffic lights, and to obtain the traffic status of and evidence about those vehicles.     

SA-PSO算法在CNN边缘提取模板设计中的应用

提出了一种用于边缘提取的细胞神经网络（CNN）模板的设计方法,该方法在基本粒子群算法的基础上引入模拟退火机制,形成模拟退火粒子群算法（SA-PSO）对模板参数值进行搜寻。在搜索过程中,用退火温度调节粒子的突跳概率,轮盘赌策略确定粒子的全局最优的替代值,这样能有效避免基本PSO算法容易陷入局部最优解的问题。同时,为了保证每轮搜寻产生的解均能使CNN网络稳定,用CNN反馈模板的研究结论对粒子群解空间进行约束。模拟实验表明,文章算法设计出的CNN模板有良好的边缘提取能力。     

Automatic image analysis and spot classification for detection of fruit fly infestation in hyperspectral images of mangoes

Fruit fly infestation of mangos is a major concern for growers and exporters, leading to requirements for quarantine treatments such as vapor heat treatment or irradiation and subsequent reduction in quality and consumer acceptance. An on-line method for detection and removal of infested fruit would thus benefit producers and consumers. An algorithm has been developed to identify spots generated in hyperspectral images of mangoes infested with fruit fly larvae. The algorithm incorporates background removal, application of a Gaussian blur, thresholding, and particle count analysis to identify locations of infestations. Each of the four algorithm steps involves adjustable parameters which were iteratively tested to find the optimal combination for detection in terms of false positive and false negative results. For algorithm parameters selected to minimize false negative results, a false negative error rate of 1.0% was achieved with 11.1% false positive error and 6.0% overall error in heavily infested samples. For the same sample set, the lowest overall error rate achieved was 2.0%, with 1.0% false positive and 3.0% false negative. For samples with lower infestation rates, the error rates were much higher, the lowest overall error being 12.3%. This therefore demonstrates the feasibility of hyperspectral imaging for fruit fly detection while highlighting the need for technology with improved resolution and signal to noise ratio to allow detection of single larvae.     

Finger gesture detection and its application to real-time control of prosthetic finger movements

A DSP-based experimental system with real-time finger joint angle detection and prosthetic finger control is designed here. The prototype system is mainly composed of four parts: real-time finger posture detection, joint angle analysis, control pulse encode, and underactuated prosthetic hand with three individually active fingers, i.e. thumb, index finger and middle finger. The real-time finger joint movement is detected by acceleration sensor ADXL330, and the microprocessor DSPTMS320F2812 is used to collect the output of acceleration sensor and calculate angle changes of the finger joint instantaneously. Then, the angle changes are encoded to PWM pulse sequence to control the stepper motor, which drove the rotation of prosthetic finger joint. Preliminary tests are given to multi-finger motion detection and real-time prosthetic finger control, respectively. The experimental results show that the acceleration sensor ADXL330 can be used to reliably detect the gesture of finger movements, it can effectively drive flexion movement and extension movement of prosthetic finger, and the grip action can be completed with the coordination of thumb, index finger, and middle finger.