Automated Insect Detection and Removal via Image Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The proliferation of insects such as thrips, mites, and aphids in food and crops, which are difficult to detect and remove due to their small size and pesticide resistance, poses a challenge in maintaining food quality and crop health.
Innovation Solution
A computer image analysis system that magnifies images of substrates to detect insects using trained image recognition software, which can trigger an action head to remove or mark detected insects, and can be mounted on movable platforms like drones for wide-area scanning and mitigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If pesticides are used to control insects, then insect population is reduced, but insect resistance develops and effectiveness decreases
Solution Approach 1:
The patent replaces chemical pesticides with an automated mechanical removal system. Image recognition software identifies insects on substrates, and robotic actuators physically remove them. This mechanical approach eliminates the biological selection pressure that causes pesticide resistance, directly addressing the contradiction between maintaining control effectiveness and preventing resistance development.
Solution Approach 2:
The system performs self-service by automatically detecting and removing insects without human intervention. The image recognition and robotic actuation work autonomously to eliminate insects, replacing the need for manual inspection and chemical application, thereby solving the reliability-effectiveness contradiction through continuous automated maintenance.
2Productivity
If washing is used to remove insects, then some insects are removed, but many insects are difficult to dislodge and remain resistant
Solution Approach 1:
The patent replaces manual washing with automated robotic actuators that can apply controlled mechanical forces to dislodge and remove insects. The robotic system can adjust its approach based on insect location and type, achieving both high productivity through automation and reliable removal completeness through adaptive mechanical action.
Solution Approach 2:
The image recognition system provides real-time feedback on insect locations and characteristics to the robotic actuators. This feedback loop enables the system to adjust its removal strategy based on observed conditions, ensuring complete and reliable insect removal while maintaining high productivity through automated operation.
3Reliability
If large scale heating is used to eliminate insects, then insects are destroyed, but the food or material to be cleansed is harmed
Solution Approach 1:
The patent replaces thermal heating with localized robotic mechanical removal. The robotic actuators physically dislodge and remove insects from the substrate without applying heat, thereby achieving reliable insect elimination while completely avoiding thermal damage to the food or material being cleaned.
Solution Approach 2:
The robotic system applies removal action only at the specific locations where insects are detected by image recognition, rather than applying widespread heating to the entire substrate. This localized approach ensures effective insect elimination only where needed, preserving the quality of the food or material throughout.
4Difficulty of detecting and measuring
If manual inspection is used to detect insects, then visible insects can be found, but small and invisible insects such as thrips, mites, and aphids remain undetected
Solution Approach 1:
The patent replaces human visual inspection with image recognition software that processes digital images at high resolution. This software can detect and identify small insects like thrips, mites, and aphids that are invisible to the naked eye, dramatically improving detection accuracy while maintaining ease of operation through automated image analysis.
Solution Approach 2:
The system creates digital copies (images) of the substrate surface and analyzes these copies through software algorithms. This copying process allows for magnification and detailed examination of the image data without physically contacting or damaging the substrate, enabling detection of microscopic insects while maintaining measurement precision.
Data Source
AI summary
A device for detecting an offending object such as weeds or insects, and addressing the offending object by removing or marking the same. The device has image acquisition and image recognition systems that captures images of plants and automatically responds by way of an action when a weed or insect is identified. The action head may grip the offending object or release a substance such a herbicide to eliminate a weed.