Fruit stem cut point positioning method, apparatus, device, and medium

By combining depth cameras and photoelectric sensors with a target detection model, precise positioning of the pineapple stem cutting point is achieved, solving the problems of low positioning accuracy and weak anti-interference ability in pineapple harvesting, improving the accuracy and safety of pineapple harvesting, and adapting to the mechanization and intelligent development of the pineapple industry.

CN122115560APending Publication Date: 2026-05-29SOUTH CHINA AGRICULTURAL UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SOUTH CHINA AGRICULTURAL UNIVERSITY
Filing Date
2026-01-26
Publication Date
2026-05-29

AI Technical Summary

Technical Problem

Existing pineapple stem cutting and positioning technologies suffer from low positioning accuracy, weak anti-interference ability, easy damage to fruits or plants, and poor adaptability, failing to meet the mechanized and intelligent harvesting needs of the pineapple industry.

Method used

By employing a depth camera combined with multiple sets of photoelectric sensors and a target detection model, and by fusing depth images with RGB images, the pineapple crown buds and fruit areas are accurately identified. Combined with motion compensation distance, the pineapple harvester is controlled to accurately reach the fruit stem cutting point.

Benefits of technology

It enables precise positioning of the pineapple stem cutting point, improves harvesting accuracy, reduces the risk of damage to the fruit and plant, and meets the needs of large-scale harvesting in the pineapple industry.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122115560A_ABST
    Figure CN122115560A_ABST
Patent Text Reader

Abstract

The application discloses a kind of fruit stem cutting point positioning method, device, equipment and medium, wherein the method is: build the detection system containing depth camera and multiple groups of photoelectric sensor, collect the RGB image and depth image of pineapple crown bud and fruit;After the RGB image is preprocessed and enhanced, the training sample is obtained by labeling target area with multi-directional rotating frame;Rotating frame-based target detection model is constructed and trained, input preprocessed RGB image, identify target area and extract rotating detection frame pixel length, combine the median of the depth value of the pixel point around the center of detection frame in depth image and conversion ratio, calculate the actual length from the top of pineapple crown bud to the connection of fruit and fruit stem;Control picker to vertically descend from above crown bud, determine the starting position of the top of crown bud by the change of photoelectric sensor detection level from high to low;Set compensation allowance, calculate the descending distance and time, and the picker stops after descending at preset speed for a certain time, and the position of the cutting mechanism at the bottom of the picker is the accurate cutting point.
Need to check novelty before this filing date? Find Prior Art