Nie Kouzhun

Bio: Nie Kouzhun is an academic researcher. The author has contributed to research in topics: Structured light & Point set registration. The author has an hindex of 3, co-authored 6 publications receiving 25 citations.

CCD camera nonlinearity degree calibration method based on structured light three-dimensional measurement system

Abstract: The invention provides a CCD camera nonlinearity degree calibration method based on a structured light three-dimensional measurement system. The method comprises the steps that according to the measurement processes of the structured light three-dimensional measurement system, on the basis that a series of standard images which are projected by a projector and accord with linear distribution are collected, a standard image gray scale-camera collected image gray scale distribution curve is drawn after certain data processing so as to further calculate the nonlinearity degree of a CCD camera; because gray scale gradient is small when the standard images are grouped, and the number of images collected by the CCD camera is large, random errors are reduced, and nonlinearity degree measurement precision is improved; the CCD camera nonlinearity degree calibration method based on the structured light three-dimensional measurement system accords with conditions of the structured light three-dimensional measurement system, and is simple in calibration process, less in middle link, and accurate in nonlinearity degree calibration result.

Method for fast recovering surface phase of high-light object based on least square method

Abstract: The invention provides a method for fast recovering a surface phase of a high-light object based on the least square method. The method is different from an original method that valid gray values and invalid gray values in N optical grating pictures are directly substituted into a standard N-step phase shifting formula to solve the phase. The method comprises the steps that firstly, invalid pixel points and invalid gray values are removed, valid gray values of valid pixel points are directly used, a linear equation set is established according to the valid gray values, the optimal solution of the equation set is solved based on the least square method, the high-precision phase value is solved according to the optimal solution, and the phase value of the surface of the high-light object can be fast recovered. The method is simple in step, high in robustness and good in random error rejection capability, extra hardware facilities are not needed, the diaphragm and exposure time of a CCD camera do not need to be adjusted, the surface of the high-light object does not need to be processed, time is saved, the operation is fast, the measuring precision of the surface of the high-light object can be guaranteed, and the measuring accuracy is high.

Method for highlight body surface phase quick recovery based on recurrence method

Abstract: The invention provides a method for highlight body surface phase quick recovery based on a recurrence method. According to the method, a method that effective gray level values in the pixel point portion are directly used is adopted instead of the prior method that effective gray level values and invalid gray level values in N raster images are directly substituted into a phase solution formula, actual values of saturated gray level values are restored through the recurrence method, and therefore the accurate phase value of a highlight body surface is recovered. The method is simple in step, high in robustness, good in random error rejection capacity, free of requirements for adjusting a CCD camera diaphragm and time of exposure and the requirement for processing the highlight body surface, capable of saving time, rapid, capable of guaranteeing measurement accuracy of the highlight body surface and high in measurement accuracy.

Space intersection and chain code similarity-based same-name mark point matching method

Abstract: The invention provides a space intersection and chain code similarity-based same-name mark point matching method The method comprises the following steps: calibrating two cameras to obtain internal and external parameters of the cameras; adhering non-coding mark points to the surface of a tested object; shooting the tested object, collecting left and right pictures, detecting and recognizing the mark points on each picture and extracting a central coordinate of the mark points; constructing a distance matrix between projection lines corresponding to the mark points on the left and right pictures according to a space intersection principle and automatically setting a rational distance threshold to realize initial matching of the mark points; constructing check chain codes for the initial matching result of the left and right pictures respectively and eliminating error matching according to the similarity of the chain codes The method is a mark point two-dimensional matching method suitable for a binocular three-dimensional visual measuring system, so that the error matching can be eliminated, the optimal threshold can be automatically set to improve the efficiency for eliminating the error matching, the matching omission is reduced, and the robustness of the algorithm is higher

Point cloud attribute compression method based on deleting 0 elements in quantisation matrix

Abstract: The invention discloses a point cloud attribute compression method based on deletion of 0 elements in a quantization matrix In view of the quantization matrix during the point cloud attribute compression process, the optimal traversal order is adopted at a coding end to enable 0 elements to be centrally distributed at a tail end in a generated data stream, entropy coding is carried out after the0 are deleted, the data volume of the data stream is reduced, a bit stream generated after coding is reduced, point cloud geometry information is combined at a decoding end to recover the deleted 0 elements, and the method is ensured not to introduce extra errors The method comprises steps: the traversal order of the quantization matrix is optimized at the coding end; the 0 elements at the tail end of the data stream are deleted; the geometry information is consulted to recover the quantization matrix at the decoding end; and a point cloud attribute compression coding process and a decoding process are carried out At the point cloud attribute compression coding end, seven traversal orders are adopted for the quantization matrix, and distribution of the 0 elements in the data stream is more centralized at the tail end; the 0 elements at the tail end of the data stream are deleted, redundant information is removed, and the data volume in need of entropy coding is reduced; and at the decoding end, the point cloud geometry information is combined to complete the deleted 0 elements, the quantization matrix is recovered according to the traversal order, and the compression performanceis improved on the premise of not introducing new errors

Depth determining method and device and electronic equipment

Abstract: The embodiment of the invention provides a depth determining method and device and electronic equipment. The method comprises the following steps: acquiring a colorful image and a depth image which are acquired by a camera and then are output; carrying out image identification based on the colorful image and determining a first image region recorded with a characteristic body in the colorful image; determining a second image region corresponding to the first image region in the depth image according to a pixel corresponding relation between the colorful image and the depth image; determining a characteristic body depth value of the second image region based on depth information corresponding to a pixel in the second image region. According to the embodiment, the characteristic body depth value of the characteristic body can be obtained under the condition that the precision of the depth image is lower and the characteristic body is difficult to identify.

High-dynamic performance three-dimensional measurement method based on adaptive grating projection

Abstract: The invention discloses a high-dynamic performance three-dimensional measurement method based on adaptive grating projection, and the method comprises the steps: setting a saturation threshold value of an image; carrying out the projection of an all-white image for a to-be-measured object, collecting the image through a camera, changing the gray scale of the projection image, generating a proper grating, and enabling the camera image not to be saturate; enabling the projection to sequentially pass through the gratings adjusted for the first time, collecting an image, solving an absolute phase through a phase shift method and a Gray code method, and completing the matching of the camera image coordinates and the projection grating image coordinates of an excessively bright region through the phase; projecting the image, collecting the image through the camera, adjusting the gray scale of the projection image pixel by pixel based on a coordinate matching result, and generating a final adaptive grating map; projecting a finally generated grating map, collecting the image, and solving an absolute phase through the phase shift method and the Gray code method; and calculating the three-dimensional coordinate information of the measured object according to a space intersection method through the calibrated camera parameters and the solved absolute phase. The method finally achieves the precise three-dimensional measurement with the high dynamic performances.

Assembly line product three-dimensional surface type detection method based on phase gradient threshold

Abstract: The invention belongs to the on-line three-dimensional surface type detection field, and especially discloses an assembly line product three-dimensional surface type detection method based on a phase gradient threshold. The invention solves the problems of the prior art, fully takes regard of the balance between measurement speed and precision required by on-line three-dimensional surface type detection, and provides an assembly line product three-dimensional surface type real time detection method based on a phase gradient threshold. The method comprises: obtaining the whole phase information of a detected object through an Fourier transform method; using a phase gradient threshold as an image characteristic to participate in correlative operation to obtain phase shift stripes; expanding a binary phase shift gradient graph to obtain an object surface abrupt change area; and finally employing the least square method to obtain the phase distribution of the abrupt change area through iteration computation, and to modify FTP calculating results in order to improve phase obtaining precision.

Binocular vision-based large-dimensional geometric quantity measurement method

Abstract: The invention provides a binocular stereoscopic vision station-transfer measurement-based large-dimensional geometric quantity measurement method. The objective of the invention is to solve problems in large-dimensional geometric quantity measurement in the manufacture field of large equipment such as airplanes and ships. The method includes the following steps that: a plane target and active vision combined calibration method is adopted to obtain visual system parameters through precise calibration, and therefore, the defect of the poor stability of the principal point of a camera caused by inaccurate positioning of an angular point can be elinminated; with a non-coding origin adopted as a measuring point, a circularity method and a fitting method are adopted to realize the identification and circular positioning of the origin, and an epipolar constraint method and a crossratio constraint method are combined to match high-density similar dots; and an SVD method is used to solve a splicing matrix between adjacent stations, and data splicing can be completed. As indicated by a test result, the relative error of the measurement result of the method and a numerical value measured by a coordinate measurement machine is smaller than 0.1%.