The invention discloses an automatic delivery method and apparatus. The method comprises steps of: constructing a delivery area map by preliminarily using binocular vision and a visual instant positioning map construction method; receiving a delivery request and obtaining cargo delivery destination information; carrying out path planning by using the delivery area map to generate path information from a delivery point to a destination; performing obstacle detection in real time by using the binocular vision during a delivery process, and dynamically correcting the delivery path to avoid detected obstacles until reaching the destination. The method can obtain an optimal path by constructing the map in real time so that a delivery robot or an unmanned aerial vehicle can flexibly cope with various scenarios, uses the optimal action plan and shortens the delivery time. The method and device realize real-time positioning and are accurate and position and high in precision.

Automatic delivery method and apparatus

In order to meeting the requirement of binocular distance-measurement at night, the invention provides a distance-measuring system and a distance-measuring method based on a shimmer binocular camera.Considering the defect of low signal noise ratio of shimmer images, the distance-measuring method comprises the following steps: performing pretreatment to reduce image noise, fully utilizing the color and texture information of global image in a feature matching and optimizing process, reducing the error matching effect caused by the local noise; combining the calculation result of the Census descriptor and color feature with a semi-global matching optimization algorithm, the robustness of feature extraction and the accuracy of feature matching are enhanced. The distance-measuring system andthe distance-measuring method have night shimmer distance-measuring ability at night and can realize the night accurate distance-measurement.

Distance-measuring system and distance-measuring method based on shimmer binocular camera

The invention discloses an automatic restaurant bill paying method based on image recognition. The automatic restaurant bill paying method comprises the steps of acquiring a current dinner plate imagethrough a first monocular camera or a first binocular camera, and acquiring feature information of each dinner plate in the current dinner plate image; comparing corresponding relationships of the feature information of each dinner plate and pre-stored dinner plate feature information and food products and prices, determining corresponding dish categories or prices, and obtaining a current user'stotal meal price by summarization. According to the method, for a canteen or restaurant using dinner plates of different shapes, dimensions and colors to place dishes of different prices, automatic image shooting of the customer order dinner plates can be achieved, dinner plate images can be automatically identified, the total prices of the dishes contained in the dinner plates can be automatic calculated, and unit price, quantity, and total price can be displayed on a display device. Therefore, the most accurate and efficient feature recognition technology can be used to identify and locatetarget objects, and the method has a very broad application prospect.

Automatic restaurant bill paying method based on image recognition

The invention discloses a depth image calculation method using inter-frame correlation of video sequences. The method comprises the following steps of: 1, correcting a picture taken by a left and right viewpoint camera; 2, for the point p of a first frame of left and right picture, calculating matching cost in a maximum parallax search range, performing parallax optimization and parallax refinement to obtain an initial parallax value d1, wherein the parallax gradient of the first frame is g1; and 3, for the point p calculation of the next frame, when the point p is not an abnormal point or an edge point, setting the parallax search range by using the parallax value of the point p of the previous frame as a reference, or calculating matching cost by using the maximum parallax search range, performing parallax optimization and parallax refinement to obtain the parallax value and the parallax gradient of the current frame for achieving real-time stereo matching. The method utilizes the parallax of the previous frame in the video sequence, greatly reduces the calculated amount of the current frame algorithm, and can shorten consumed time and improve efficiency in a practical application scene under the condition that the matching precision of an original algorithm is guaranteed.

Depth image calculation method using inter-frame correlation of video sequences

The invention provides a 3D tooth imaging modeling method. The method comprises the steps: obtaining a tooth image of an oral cavity through a binocular camera; carrying out the denoising filtering of the tooth image; calibrating the binocular camera, and obtaining a reprojection matrix; calculating three-dimensional coordinates of the tooth image through the reprojection matrix and the tooth image after noise reduction, and obtaining the point cloud data; and carrying out the curved surface reconstruction of the point cloud data through a space Poisson equation. Compared with a conventional method, the method can quickly obtain a tooth model, is low in cost, does not need to employ a conventional die obtaining material, reduces the waste, and can improve the treatment experience of a patient.

3D tooth imaging modeling method

The invention discloses a stereo matching three-dimensional reconstruction method based on dynamic programming. A system according to the method is composed of two video cameras. The stereo matching three-dimensional reconstruction method comprises the following steps of (1), adjusting the positions of the two video cameras so that imaging planes of the two video cameras are parallel as possible; (2), performing calibration on a three-dimensional measuring system, obtaining inner parameters and outer parameters of the two video cameras, and obtaining a correspondence between pixel coordinates on an image and a world coordinate system; (3), performing epipolar line rectification and image conversion; (4), obtaining a parallax graph by means of a stereo matching algorithm based on dynamic programming; (5), performing parallax correction; and (6), obtaining a three-dimensional point cloud according to a video camera calibration parameter and the parallax graph through a spatial intersection method. The stereo matching three-dimensional reconstruction method has advantages of high parallax graph precision and high real-time performance. Furthermore the three-dimensional point cloud of the image can be accurately, quickly and automatically reconstructed.

Da Feipeng

Papers

Stereo matching three-dimensional reconstruction method based on dynamic programming