Search

• A multi-camera differential binocular vision sensor and its measurement principle are proposed, and the contradiction between FOV and measurement is overcome. • The structure parameters of the sensor are optimized from the aspects of spatial arrangement, measurement range, and accuracy to realize high-precision 3D measurement. • Experiments have proved that the sensor expands the FOV of traditional binocular cameras, and has higher measurement accuracy than other visual measurement methods. The contradiction between the field-of-view (FOV) and measurement accuracy is an urgent problem to be solved in the field of vision precision measurement. The binocular vision sensor with two cameras can theoretically ensure high-precision measurement, but field stitching is carried out to expand FOV resulting in a decrease in measurement accuracy. To ensure high-precision measurement while expanding the FOV, this paper proposes a multi-camera differential binocular vision sensor. Four auxiliary cameras are distributed around the central main camera, and the four quadrant images of the main camera form four pairs of binoculars with the four auxiliary cameras. The structure parameters of the sensor are optimized from the aspects of spatial arrangement, measurement range, and accuracy. Compared to traditional binocular vision sensors, multi-camera differential binocular vision sensor has wide FOV and high measurement accuracy, which is suitable for applications in areas such as intelligent driving, robot navigation, and motion analysis. [ABSTRACT FROM AUTHOR]

In this paper, a phase based method to calibrate a binocular vision sensor is presented. In this method, only a surface plate is utilized. A series of phase shifting patterns are projected onto the surface plate. In this case, fundamental matrix can be calculated out from point correspondences on two image planes. As intrinsic parameters of each camera have been obtained according to related calibration method, essential matrix can be worked out. Then, rotation matrix and translation matrix with a coefficient are deduced based on singular value decomposition of the essential matrix. As size of the pattern can be confirmed from a look-up table, the coefficient is determined. So far, the binocular vision sensor is calibrated. In our proposed method, only one planar target without any pattern is necessary, which is with high precision and easy machined. As the calibration method is based on related phase algorithm, the calibration process is not affected by the ambient light which makes our calibration method robust. Experiment results show the precision of our proposed method. When we utilized a laser displacement sensor with a precision of 0.01 mm to determine size of the pattern, root mean square error of our calibration method is 0.103 mm with regard to the measurement area of about $100\times 80$ mm. Moreover, as planar features are widely existing, our proposed method is very suitable for on-site calibration and auto-calibration.

The presented movable vision measurement for the three-dimensional (3D) surface of a large-sized object has the advantages of system simplicity, low cost, and high accuracy. Aiming at addressing the problems of existing movable vision measurement methods, a more suitable method for large-sized products on industrial sites is introduced in this paper. A raster binocular vision sensor and a wide-field camera are combined to form a 3D scanning sensor. During measurement, several planar targets are placed around the object to be measured. With the planar target as an intermediary, the local 3D data measured by the scanning sensor are integrated into the global coordinate system. The effectiveness of the proposed method is verified through physical experiments.

Most of the existing calibration methods for binocular stereo vision sensor (BSVS) depend on a high-accuracy target with feature points that are difficult and costly to manufacture. In complex light conditions, optical filters are used for BSVS, but they affect imaging quality. Hence, the use of a high-accuracy target with certain-sized feature points for calibration is not feasible under such complex conditions. To solve these problems, a calibration method based on unknown-sized elliptical stripe images is proposed. With known intrinsic parameters, the proposed method adopts the elliptical stripes located on the parallel planes as a medium to calibrate BSVS online. In comparison with the common calibration methods, the proposed method avoids utilizing high-accuracy target with certain-sized feature points. Therefore, the proposed method is not only easy to implement but is a realistic method for the calibration of BSVS with optical filter. Changing the size of elliptical curves projected on the target solves the difficulty of applying the proposed method in different fields of view and distances. Simulative and physical experiments are conducted to validate the efficiency of the proposed method. When the field of view is approximately 400 mm × 300 mm, the proposed method can reach a calibration accuracy of 0.03 mm, which is comparable with that of Zhang's method. [ABSTRACT FROM AUTHOR]

Most of the existing calibration methods for binocular stereo vision sensor (BSVS) depend on a high-accuracy target with feature points that are difficult and costly to manufacture and. In complex light conditions, optical filters are used for BSVS, but they affect imaging quality. Hence, the use of a high-accuracy target with certain-sized feature points for calibration is not feasible under such complex conditions. To solve these problems, a calibration method based on unknown-sized elliptical stripe images is proposed. With known intrinsic parameters, the proposed method adopts the elliptical stripes located on the parallel planes as a medium to calibrate BSVS online. In comparison with the common calibration methods, the proposed method avoids utilizing high-accuracy target with certain-sized feature points. Therefore, the proposed method is not only easy to implement but is a realistic method for the calibration of BSVS with optical filter. Changing the size of elliptical curves projected on the target solves the difficulty of applying the proposed method in different fields of view and distances. Simulative and physical experiments are conducted to validate the efficiency of the proposed method. When the field of view is approximately 400 mm × 300 mm, the proposed method can reach a calibration accuracy of 0.03 mm, which is comparable with that of Zhang's method., Competing Interests: The authors declare no conflict of interest.