Ripe fruit machine identification method
A technology of machine recognition and fruit, applied in character and pattern recognition, instruments, computer parts, etc., can solve the problems of limited promotion, lighting conditions and image acquisition angle, and achieve the effect of small impact
Active Publication Date: 2021-12-10
HUNAN BIOLOGICAL & ELECTROMECHANICAL POLYTECHNIC
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] Machine recognition of fruits on trees mainly includes three steps: extraction of unoccluded fruit contour segments, restoration of fruit contours, and calculation of fruit position and size parameters. The previous methods are greatly affected by lighting conditions and image acquisition angles in the contour extraction process. cause limited promotion
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0041] Carry out the machine identification of the fruit on the mature tree according to the following methods and steps:
[0042] Step A: Collect the fruit tree image f with fruit 0 (x,y), such as figure 1 shown. In the orchard, the fruit tree image collected by the vision system of the fruit picking robot f 0 (x,y) is an RGB color image.
[0043] Step B:
[0044] B1. to f 0 (x, y) is binarized, and the RGB color image f 0 (x,y) is converted into a black and white binary image f 1 (x,y), such as figure 2 shown. in, In the formula, is the average color tone of ripe fruit, T is the threshold value, and T is determined by the fruit tree image f 0 The γ component of the fruit area and non-fruit area of (x, y) is statistically determined; R (x, y), G (x, y), and B (x, y) are the RGB color image f 0 Red, green, and blue components of (x,y);
[0045] B2. In the black and white binary image f 1 In (x, y), due to the complexity of the background and the influence of...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a method for machine recognition of ripe fruit, comprising: collecting fruit tree images; converting the fruit tree images into black and white binary images and removing false background color blocks and fruit false targets; obtaining fruit contour coordinate matrix; forming a clockwise arrangement The fruit contour coordinate set; form the fruit contour intersection set; select the unoccluded fruit contour in each connected area, the unoccluded ellipse arc segment in the partially occluded fruit contour; the selected unoccluded fruit contour, partially occluded The least squares fitting is performed on the unoccluded elliptical arc segment in the fruit contour, and the false fruit target whose size is 0.3 to 1.7 times larger than the normal fruit is discarded; the center coordinates and radius of the circular contour curve are calculated. The invention is less affected by the lighting conditions, and is not sensitive to the angles of taking pictures such as front light, backlight, side light, etc., ensures the accuracy of the outline extraction of the blocked fruit area, and can detect partially blocked and non-blocked fruits at the same time, and the detection time is short. High accuracy.
Description
technical field [0001] The invention particularly relates to a method for machine identification of ripe fruit. Background technique [0002] The real-time online identification of fruits on trees is one of the core technologies of fruit harvesting robots, and it is also a basic problem of fruit information production management in plantations. The rapid identification of occluded fruits is the first difficult problem to be overcome in these problems. [0003] Machine recognition of fruits on trees mainly includes three steps: extraction of unoccluded fruit contour segments, restoration of fruit contours, and calculation of fruit position and size parameters. The previous methods are greatly affected by lighting conditions and image acquisition angles in the contour extraction process. resulting in limited promotion. Contents of the invention [0004] The purpose of the present invention is to provide a method for machine recognition of ripe fruit in view of the deficienc...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/38G06K9/46
CPCG06V10/28G06V10/44G06V20/68
Inventor 曹乐平
Owner HUNAN BIOLOGICAL & ELECTROMECHANICAL POLYTECHNIC