Threshing State Management Using Neural Network Image Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current threshing control systems for combine harvesters face challenges in effectively managing crop loss due to the reliance on a large number of sensors and instantaneous loss measurements, which can lead to inadequate control and increased threshing loss, especially when dealing with varying threshing processing amounts and mixed foreign matter.
Innovation Solution
A threshing state management system utilizing an image capture unit, state detection neural network, and parameter determination unit to accurately estimate the threshing processing state and control the threshing apparatus, reducing crop loss by adjusting parameters such as sieve opening and vehicle speed based on real-time image data and travel state inputs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a large number of sensors and instantaneous loss measurements are used, then measurement precision is improved, but device complexity increases and reliability decreases due to inadequate control
Solution Approach 1:
The patent replaces the mechanical sensor-based measurement system with an image processing system. Instead of using multiple physical sensors to detect threshing loss, the system uses cameras to capture images of the threshing process and employs image processing algorithms to analyze and quantify loss. This substitution reduces device complexity while maintaining or improving measurement precision through computational analysis rather than physical sensing.
Solution Approach 2:
The patent introduces an intermediary image processing system between the threshing apparatus and the control system. Rather than directly measuring loss with sensors, the system captures visual information through images and processes this intermediate data to derive loss information. This intermediary approach simplifies the overall system architecture while providing comprehensive loss measurement capabilities.
2Loss of time
If instantaneous loss measurements are used, then response time is improved, but measurement precision deteriorates due to inadequate control under varying processing amounts
Solution Approach 1:
The patent implements continuous image capture and processing during the threshing operation. Instead of intermittent sensing, the system continuously monitors the threshing process through ongoing image acquisition and analysis. This continuous action enables real-time detection of loss variations while adapting to changing processing conditions, maintaining both rapid response and accurate measurement throughout the operation.
Solution Approach 2:
The patent employs dynamic image processing that adapts to varying threshing conditions. The system adjusts its analysis parameters and processing methods based on real-time observations of the threshing process, allowing it to maintain measurement precision despite changes in processing amount, crop type, or operational conditions. This dynamic adaptation enables accurate loss measurement across diverse operating scenarios.
3Device complexity
If traditional sensor-based control is used, then device complexity is reduced, but productivity decreases due to increased threshing loss
Solution Approach 1:
The patent implements a feedback control system where image processing results are continuously fed back to adjust threshing apparatus parameters. The system analyzes images of the threshing process, quantifies loss, and automatically adjusts operational parameters such as threshing drum speed, sieve opening, or winnower angle. This closed-loop feedback mechanism reduces threshing loss and improves harvesting efficiency while maintaining manageable system complexity through automated control.
Solution Approach 2:
The patent utilizes parameter change optimization by adjusting key operational parameters based on image analysis results. The system modifies parameters such as threshing intensity, sieve aperture size, or material flow rates in response to detected loss conditions. By dynamically optimizing these parameters, the system maximizes harvesting efficiency and minimizes grain loss without requiring overly complex control architectures.
Data Source
AI summary
A threshing state management system includes an image capture unit 80 that captures an image of a threshed material threshed by a threshing apparatus, a state detection neural network 72 that outputs a threshing processing state in the threshing apparatus based on image input data generated based on the captured image from the image capture unit 80, a parameter determination unit 73 that determines a control parameter of the threshing apparatus based on the threshing processing state, and a threshing control unit TU that controls the threshing apparatus based on the control parameter.