Hybrid Vision Crop Navigation With Selective Image Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current agricultural technologies face inefficiencies and high costs in weed remediation and crop detection, as existing methods rely on manual or chemical means and struggle to accurately distinguish between plants and weeds, especially due to varying growth stages and lack of contrast in digital images.
Innovation Solution
An autonomous vehicle system equipped with GPS, high-resolution cameras, and machine vision capabilities that selectively switches between low-resolution and high-resolution image capture modes for rapid field traversal and accurate plant classification, using field maps and geolocation data to differentiate between crops and weeds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual methods with physical removal or herbicide spraying are used for weed remediation, then weed control can be achieved, but labor costs and fuel consumption increase significantly
Solution Approach 1:
The system uses autonomous vehicles equipped with machine vision to automatically detect and classify plants and weeds, enabling the system to serve itself without human intervention in field traversal and inspection tasks
Solution Approach 2:
The patent replaces manual mechanical weed removal and herbicide spraying with an autonomous vision-based system that uses machine learning algorithms to identify and locate weeds, substituting mechanical and chemical methods with intelligent detection and navigation
2Extent of automation
If computer vision techniques are used to detect plants and weeds, then automation is improved, but the ability to distinguish between different objects remains insufficient
Solution Approach 1:
The system changes the parameters of image analysis by using multiple contrast value ranges and adapting detection thresholds based on growth stage, allowing the same vision system to accurately detect both early-stage seedlings and mature plants with different visual characteristics
Solution Approach 2:
The patent implements dynamic adjustment of detection parameters based on the growth stage of plants, making the vision system adaptive rather than static, so it can distinguish between crops and weeds at different developmental phases
3Productivity
If image recognition relies on specific contrast values, then detection can be performed, but accuracy decreases during early growth phases when contrast is inadequate
Solution Approach 1:
The system dynamically changes contrast value parameters based on the detected growth stage, using lower contrast thresholds for early seedling detection and higher thresholds for mature plant identification, thereby maintaining accuracy across all growth phases
4Measurement precision
If high-resolution image capture is used for accurate plant classification, then detection precision is improved, but traversal speed and efficiency decrease
Solution Approach 1:
The patent implements a dynamic resolution switching system that adjusts image capture quality based on whether the vehicle is in navigation mode or classification mode, optimizing the trade-off between speed and precision at different operational phases
Solution Approach 2:
The system segments the image processing task into two stages: rapid low-resolution scanning for navigation and location identification, followed by high-resolution capture only at specific points for detailed classification, thereby maintaining overall efficiency while achieving accurate identification
Data Source
AI summary
Autonomous vehicles with global positioning systems are used for field inspection. A vehicle may be programmed to traverse a field, while using sensors to detect objects in the field, and then capture low-resolution images of the objects. Machine vision techniques are used with the low-resolution images to recognize the objects as crops, non-crop plant material or undefined objects. Location data is used to correlate recognized objects with digitally stored field maps to resolve whether a particular object is in a location at which crop planting is expected or not expected. Depending on whether an object in a low-resolution digital image is recognized as a crop, and whether the object is in an expected geo-location for crops, the vehicle may switch to a second image capture mode, for example, capturing a high-resolution image of the object, and/or execute a disease analysis and/or weed analysis on the images of the objects.