Vision-Guided Crop Leaf Removal System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for removing crop leaves, such as those described in NL 102 47 02 and NL 2000333, are inefficient and complex, requiring substantial software and mechanical complexity, making them less competitive with manual removal methods, especially when trying to remove leaves from beneath fruits or in dense crops.
Innovation Solution
A method using vision techniques to observe crops from a low position, determining the lowest leaf to be removed, and using stripping means to separate it from the central stem while collecting or transporting the leaf away, allowing for unobstructed observation and removal of subsequent leaves, with the option to start from beneath or above and differentiate between leaves and fruits.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a robot arm approaches the main stem of the crop and uses cutting blades to remove leaves, then leaves can be removed from the crop, but the stalks are not cut through in the best possible place
Solution Approach 1:
The patent replaces the mechanical cutting blade system with a vision-guided robotic arm system that can precisely locate and remove leaves at optimal positions. The vision system captures images of the crop, identifies leaf positions and orientations, and guides the robotic arm to achieve precise cutting locations without the mechanical complexity of adjustable blades.
Solution Approach 2:
The patent changes the operational parameters by transitioning from fixed mechanical cutting heights to dynamically adjustable cutting positions based on vision system detection. The system determines the optimal cutting position for each leaf based on its detected orientation and position relative to the main stem, allowing precise control over where each leaf is removed.
2Extent of automation
If complex software and force-sensing cutting arms are used to remove crop parts, then cutting decisions can be made, but the number of parts removed per unit time is insufficient to be competitive with manual removal
Solution Approach 1:
The patent replaces complex force-sensing mechanical systems with a vision-based automated system. The vision system rapidly captures and processes images to identify leaves for removal, eliminating the need for complex force-sensing feedback loops and enabling faster automated decision-making about which leaves to remove and where to cut.
Solution Approach 2:
The patent uses vision techniques to create a digital representation (image capture) of the crop structure, allowing the system to analyze and plan leaf removal without physical contact. This optical copying approach enables rapid assessment and decision-making that is faster than manual inspection while avoiding the complexity of force-sensing systems.
3Difficulty of detecting and measuring
If a camera is moved into the dense crop to observe leaves for removal, then leaves can be identified, but the view is obstructed and observation is difficult
Solution Approach 1:
The patent changes the observation dimension by positioning the camera at an angle relative to the crop rows rather than directly within the dense crop canopy. This angular positioning provides an unobstructed view of the leaves and main stems while maintaining the ability to identify which leaves should be removed, eliminating the need to physically move the camera into dense vegetation.
4Productivity
If leaves are stripped from the crop, then leaf removal is achieved, but the stripped leaves must be removed from the vicinity of the crop to allow observation of the next leaf
Solution Approach 1:
The patent ensures continuous operation by integrating leaf removal with the camera's observation cycle. As the robotic arm removes a leaf, the camera immediately captures the next image to identify the subsequent leaf for removal. This continuous cycle of observation and removal eliminates idle time and maintains steady productivity without requiring separate operations for leaf removal and re-positioning for the next observation.
Data Source
AI summary
Method and device for the removal of a part of a crop, such as a leaf (14). To this end, the crop is approached from a low position with vision techniques and the stem (12) and the parts protruding therefrom are observed from beneath. Based upon the number of images observed, an arm is controlled and moved towards the relevant stalk (13). This movement is primarily parallel to the stalk and is performed from a low proximity position. When the stalk (13) is approached, the stalk (13) is positioned within an opening between two rotating parts (3). The stalk is grasped by way of rotation and the stalk is moved in respect of the arm so that the cutting point of the stalk is manipulated towards the arm. The stalk (13) is subsequently cut through and the leaf is disposed of.