Radial base neural net-based infrared photoelectric passenger flow collecting device and method

The invention relates to an infrared photoelectric flow collecting device and the method based on a radial-basis function neural network, combining the infrared photoelectric detecting technology and the radial-basis function neural networks to build a flow number system based on the RBF neural networks which is used for the flow collecting in public place. Four groups of the Opposite-type Photoelectric Sensor are arranged at the gateway of the market and at the position equaling to the height of the ankle, connecting with a computer through a switch interface card, and recording a variational signal generated by the customers' sheltering the sensors in a base address in the computer when the customers pass through the infrared sensors district. The data in the base address is stored by adopting an effective data storage structure, a pretreatment, a partition and a characteristic extracting are carried out to the data, the results are regarded as the input of the radial-basis function neural networks, so that the number of customers continuously passing through the infrared sensors district is accurately identified. The method enhances the accuracy of the real-time customer count, and can identify the condition of more people side-by-side, the error rate is low.