Camera-Based Threshing Control for Unthreshed Grain Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing grain harvesting systems struggle to accurately assess the effectiveness of threshing operations due to the difficulty in distinguishing small grain kernels from larger non-grain crop material, leading to unreliable grain loss detection and ineffective threshing adjustments.
Innovation Solution
Implement a camera system to capture images of the crop flow downstream of the threshing system, analyze grain ears for physical properties such as unthreshed or partially unthreshed status, and adjust operational settings based on these properties to optimize threshing intensity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If impact sensors are used downstream of the threshing system to detect grain loss, then grain loss detection capability is provided, but the high volume and big pieces of non-grain crop material make the sensors far less reliable
Solution Approach 1:
The patent replaces mechanical impact sensors with an optical imaging system (camera) to detect grain kernels. The camera captures images of the crop flow downstream of the threshing system, and image processing algorithms identify and count grain kernels based on their visual characteristics, eliminating the reliability issues of mechanical sensors in high-volume non-grain material environments
Solution Approach 2:
The patent creates an optical copy (image) of the grain kernels and non-grain material instead of using physical contact sensors. By capturing visual information and processing it through algorithms, the system replicates the detection function without the mechanical components that fail in high-volume material streams
2Measurement precision
If image sensors are used to distinguish grain kernels in the crop flow, then grain detection capability is provided, but it is difficult to recognize small grain kernels in the presence of high volume of much bigger pieces of non-grain crop material
Solution Approach 1:
The patent applies local quality by using specific image processing techniques focused on the visual characteristics of grain kernels (size, shape, color, texture) to distinguish them from non-grain material. The algorithm analyzes local features of each detected object in the image stream, allowing accurate identification of small grain kernels even when surrounded by much larger non-grain pieces
Solution Approach 2:
The patent changes parameters by using multiple image processing parameters (grain kernel size thresholds, color ranges, shape factors, texture patterns) to differentiate grain from non-grain material. By adjusting and combining multiple detection parameters, the system overcomes the difficulty of recognizing small grain kernels against a background of varied non-grain material
3Productivity
If threshing action is increased to detach grain from crop materials, then grain separation effectiveness is improved, but excessive vegetative material covers too many apertures causing grain loss
Solution Approach 1:
The patent implements feedback by using the camera system to continuously monitor the crop flow downstream of the threshing system. The detected grain kernel count and characteristics are fed back to the control system, which automatically adjusts threshing parameters (rotor speed, concave clearance) to optimize grain separation while preventing excessive material buildup that would block apertures and cause grain loss
4Loss of substance
If the combine ground speed is reduced to decrease threshing system loading, then grain loss is reduced, but productivity decreases
Solution Approach 1:
The patent applies dynamics by enabling real-time, continuous adjustment of threshing system parameters based on actual operating conditions monitored by the camera system. Instead of static speed reduction, the control system dynamically adjusts rotor speed, concave clearance, and other parameters to maintain optimal grain separation across varying loading conditions, allowing the combine to operate at full ground speed without grain loss
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method is provided for controlling a threshing system (24) for an agricultural harvester (10). The method comprises using a camera (80, 81) for obtaining images of a crop flow, processing the obtained images and controlling an operational setting of the threshing system (24). The images are obtained downstream of the threshing system (24), preferably somewhere between the threshing rotor (40) or threshing drum and the residue spreader (74). The image processing aims at detecting grain ears in the obtained images, and to derive from those images at least one physical property of the detected grain ears. The operational setting of the threshing system (24) are controlled in dependence of the derived at least one physical property.