Road Surface Reconstruction by Stereo Vision

Abstract

This paper covers the problem of road surface reconstruction by stereo vision with cameras placed behind the windshield of a moving vehicle. An algorithm was developed that employs a plane-sweep approach and uses semi-global matching for optimization. Different similarity measures were evaluated for the task of matching pixels, namely mutual information, background subtraction by bilateral filtering, and Census. The chosen sweeping direction is the plane normal of the mean road surface. Since the cameras’ position in relation to the base plane is continuously changing due to the suspension of the vehicle, the search for the base plane was integrated into the stereo algorithm. Experiments were conducted for different types of pavement and different lighting conditions. Results are presented for the target application of road surface reconstruction, and they show high correspondence to laser scan reference measurements. The method handles motion blur well, and elevation maps are reconstructed on a millimeter-scale, while images are captured at driving speed.

A New Vision Measurement Technique with Large Field of View and High Resolution

“…Based on Equation (20), one can intuitively form the following conclusions: 1. The resolution of the parallel optical axis vision system is proportional to the sampling interval.…”

“…However, most researchers of visual technology tend to focus on improving the measurement accuracy and success rate by investigating (i) different calibration algorithms [ 7 , 8 , 9 , 10 , 11 ] to improve the calibration accuracy, (ii) different matching algorithms to improve the matching accuracy [ 12 , 13 , 14 , 15 ], and (iii) three-dimensional reconstruction algorithms to improve the reconstruction accuracy [ 16 , 17 , 18 ]. In addition, researchers have focused on expanding the application scenarios [ 19 , 20 ]. Some researchers have studied the relationship between visual system structural parameters and measurement accuracy [ 21 , 22 , 23 , 24 , 25 ].…”

Đề xuất thuật toán ước lượng độ cao bờ sông sử dụng máy ảnh ba chiều

“…Nhiều phương pháp thu thập dữ liệu độ cao ứng dụng máy ảnh đã được các nhà khoa học đề xuất. Trong đó, hệ thống hai máy ảnh là giải pháp được quan tâm nhất và được áp dụng trong nhiều ứng dụng như: đánh giá hiện trạng mặt đường (Brunken, 2021), đo sóng hay xác định hướng sóng dựa vào độ cao mặt nước (Molfetta et al, 2020;Vieira et al, 2020) và nhiều ứng dụng dựa vào độ cao bề mặt khác nữa. Các nghiên cứu này đã chứng minh được tính hiệu quả, nhưng để có thể áp dụng cần am hiểu các thuật toán xử lý ảnh hay thị giác máy tính.…”

Detection and Analysis of Pavement-Section Based on Laser Displacement Sensor

“…The detection of Pavement-Sections includes longitudinal-section detection and cross-section detection [ 6 ]. Pavement profile detection is the detection of the vertical plane cut along the longitudinal direction of the road [ 7 ], which mainly reflects the road surface undulation of the road in the medium and small wavelengths, and is used to detect the smoothness of the road surface [ 8 ]; Pavement cross-section detection is a plane detection perpendicular to the longitudinal-section of the road [ 9 ], which mainly reflects the lateral undulation of the road. It is used to detect the rutting and structural depth of the road surface [ 10 ] and can also detect the lateral slope of the road surface [ 11 ].…”