3D Vision Measurement for Small Devices Based on Consumer Sensors

High-precision, low-cost three-dimensional (3D) space measurement and positioning technology is desperately needed in wide applications. This study analyses the key technologies in the recognition of the devices to achieve the requirement of device recognition and capture on the production line. A 3D measurement algorithm for the small devices based on the consumer-level sensors is proposed in this study. Histogram of gradients feature is used to classify the devices, and structure light is used to get the depth data of the devices. Object extraction and Euclidean cluster segmentation are used to analyse the depth data, in order to determine their positions and orientations. In the database built on iPhone X, the accuracy of category identification reached 0.97, and the measurement error of angle is small. The results show that the proposed method is feasible and can be applied to the recognition and position of the devices.