Method for judging interference type of bunch-type fruit mother branch based on visual scaling method

Abstract

The invention provides a method for judging the interference type of a bunch type fruit mother branch based on a visual scaling method, and the method comprises the steps: firstly achieving the pre-positioning of the mother branch through the cooperation of a monocular camera and a binocular camera, and obtaining a pre-positioning image and a pre-positioning geometric center point of the mother branch; then obtaining an actual positioning image and an actual positioning geometric center point of the mother branch in the picking process through a binocular vision camera; and finally, comparing the actual positioning image and the pre-positioning image of the mother branch by using a monocular vision zooming method, and determining the interference type of the mother branch. By means of the method, the fruit bunches can be accurately and effectively recognized, the mother branches of the fruit bunches can be pre-positioned, the interference types of the mother branches can be accurately judged in the picking process, influences of external factors are avoided, then the picking robot accurately cuts off the mother branches of the fruit bunches, the integrity of the fruit bunches is guaranteed, the picking efficiency is improved, the picking cost is reduced, and labor is saved.

Description

technical field

[0001] The invention relates to the technical field of fruit intelligent picking, in particular to a method for judging the interference type of string-shaped fruit branches based on a visual zoom method. Background technique

[0002] my country is an important producer of string fruits such as longan, grapes, and lychees, and is also deeply loved by consumers; for example, lychees are distributed in the southwest, south, and southeast of my country, and Guangdong and southern Fujian are the most cultivated. Bananas, pineapple, and longan are known as the "four major fruits in the southern country".

[0003] For the picking of bunch-shaped fruits, it mainly relies on manual work at present. However, manual picking is labor-intensive and expensive, and wastes a lot of manpower and material resources. Second, manual picking is inefficient, time-consuming, and costly.

[0004] With the advancement of science and technology, mechanical automation and intelligence...

Examples

Embodiment

[0054] Such as Figure 1~3 As shown, taking litchi as an example, a method for judging the interference type of string-shaped fruit branches based on the visual zoom method is characterized in that:

[0055] S100, first realize the pre-positioning of the mother branch through the cooperation of the monocular CCD camera and the binocular CCD camera, and the specific steps are:

[0056] S101: Use a monocular CCD camera to randomly acquire multiple color images including string-shaped fruits, leaves, and branches; select and divide multiple fruit objects and non-fruit objects in the color image, and extract fruit objects and non-fruit objects respectively The texture feature value and color feature value of the target are used as positive and negative samples;

[0057] S102. Use the support vector machine SVM to train the positive and negative samples to generate multiple weak classifiers; then use the AdaBoost algorithm to construct a strong classifier, use the strong classifie...

technical field

[0001] The invention relates to the technical field of fruit intelligent picking, in particular to a method for judging the interference type of string-shaped fruit branches based on a visual zoom method. Background technique

[0002] my country is an important producer of string fruits such as longan, grapes, and lychees, and is also deeply loved by consumers; for example, lychees are distributed in the southwest, south, and southeast of my country, and Guangdong and southern Fujian are the most cultivated. Bananas, pineapple, and longan are known as the "four major fruits in the southern country".

[0003] For the picking of bunch-shaped fruits, it mainly relies on manual work at present. However, manual picking is labor-intensive and expensive, and wastes a lot of manpower and material resources. Second, manual picking is inefficient, time-consuming, and costly.

[0004] With the advancement of science and technology, mechanical automation and intelligence...

Examples

Embodiment

[0054] Such as Figure 1~3 As shown, taking litchi as an example, a method for judging the interference type of string-shaped fruit branches based on the visual zoom method is characterized in that:

[0055] S100, first realize the pre-positioning of the mother branch through the cooperation of the monocular CCD camera and the binocular CCD camera, and the specific steps are:

[0056] S101: Use a monocular CCD camera to randomly acquire multiple color images including string-shaped fruits, leaves, and branches; select and divide multiple fruit objects and non-fruit objects in the color image, and extract fruit objects and non-fruit objects respectively The texture feature value and color feature value of the target are used as positive and negative samples;

[0057] S102. Use the support vector machine SVM to train the positive and negative samples to generate multiple weak classifiers; then use the AdaBoost algorithm to construct a strong classifier, use the strong classifie...

Images