Accurate detection method and system for dense plums picked by robot

Abstract

The invention discloses a method and a system for accurately detecting dense plums picked by a robot. The method comprises the following steps: acquiring images of fruits in an orchard through image acquisition equipment; performing inspection processing to obtain a target detection image meeting requirements; performing data annotation of fruits with different maturity degrees on the target detection image to obtain an annotated image; dividing the marked image into a training set and a test set according to a proportion, and obtaining the number of mature proportion fruits in the training set; judging whether data balance processing is needed or not according to the fruit ripening proportion in the training set; performing data enhancement processing on the balance training set data; improving the target detection model; training and predicting the data enhancement training set by improving the target detection model to obtain a detection result; according to the invention, the deep learning model is applied to identification and picking of plums, can be deployed on a robot picking platform, and provides technical support for yield estimation of orchards and research of picking robots.

Description

technical field [0001] The invention relates to the research field of agricultural robots, in particular to a method and system for precise detection of dense plums for robot picking. Background technique [0002] Compared with apples, citrus, mango and other fruits, plums are small and densely distributed, and are easily blocked by fruits or branches and leaves. Most plum trees are planted on hillsides, and the growth environment of their fruits is full of complexity and uncertainty. In the current plum orchard, the tasks of identifying the maturity of plums and picking plums are all completed by the fruit growers. Today, labor costs have risen greatly, and the proportion of labor costs in the total cost is also increasing. According to the survey, the increase in labor costs in 2019 is as high as 12-15%. In precision agriculture, labor shortages and an aging population add resistance to the development of the fruit industry. To sum up, the mechanization and intelligenc...

AI Technical Summary

Problems solved by technology

Although relevant scholars have done research work on the detection of plums, the characteristics of clustered growth and mixed maturity of plums make it more difficult to detect small targets, which leads to the poor performance of existing algorithms in the process of plum detection. performance

So far, there has been no research on deep learning methods for detecting dense plums in natural environments, resulting in limited technical resources for fruit picking robots in orchards

Images