Optical Flow Crop Jam Detection in Harvesters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current self-propelled harvesting machines require manual visual monitoring of crop flow, leading to inefficiencies and increased labor costs due to the difficulty in detecting impending crop jams, as existing sensor systems struggle with accurate and timely detection of irregularities in crop flow.
Innovation Solution
An agricultural working machine equipped with an image processing system that determines optical flow and vector fields from preprocessed images, allowing for early detection of crop jams by calculating crop flow speed profiles, thereby simplifying the monitoring and control of crop flow without the need for complex feature identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual visual monitoring of crop flow is used, then the operator can detect crop flow issues, but the detection is delayed and labor-intensive
Solution Approach 1:
The patent replaces manual visual monitoring with an automated optical sensing system that uses cameras and image processing algorithms to detect crop flow irregularities. The sensor system captures images of the crop flow and automatically analyzes them to identify potential jams, eliminating the time delay and labor associated with manual monitoring while improving detection accuracy through continuous automated analysis.
Solution Approach 2:
The patent introduces an image processing system as an intermediary between the crop flow and the operator. This system processes images from sensors, identifies irregularities through algorithmic analysis, and provides early warnings before actual jams occur. The intermediary system enables timely detection by continuously analyzing crop flow characteristics and alerting operators to potential issues before they develop into full blockages.
2Measurement precision
If complex image recognition methods are used to detect crop characteristics, then detection accuracy improves, but system complexity and processing requirements increase
Solution Approach 1:
The patent extracts and focuses on specific, easily identifiable image features such as motion blur patterns and optical flow vectors that directly indicate crop flow irregularities. By selecting only the most relevant features for detection rather than attempting comprehensive crop characteristic analysis, the system achieves accurate jam detection while maintaining relatively simple processing requirements and avoiding unnecessary complexity.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables continuous, area-wide determination of crop flow velocities, allowing for timely detection of impending jams and automatic control measures to prevent crop jams, reducing labor and increasing harvesting efficiency.
Implementation Method 1
at least one sensor system associated with the crop recovery device for detecting a crop flow in the crop recovery device, wherein the sensor system is coupled for transmitting the images to the image processing system
Implementation Method 2
In a further step, an optical flow and corresponding vector fields are determined from the preprocessed images, and in a subsequent step, the crop flow velocity profiles are derived and evaluated from the corresponding vector fields
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to an agricultural machine (1), in particular a self-propelled harvester, comprising a crop recovery device (3) comprising a crop cutting device (10), a crop receiving device (8) and/or a crop intake device, and at least one crop conveying device (14), and an image processing system (20), a data output unit, and at least one sensor system (17) associated with the crop recovery device (3) for detecting a crop flow (22) in the crop recovery device (3), wherein the sensor system (17) is coupled to the image processing system (20) for transmitting the images (23), wherein the images generated by the sensor system are transferred to the image processing system, wherein the image processing system (20) processes a selection of available images (23), and wherein the image processing system (20) preprocesses the transmitted images (23) in a first step (S1).In a further step (S2), an optical flow (25) and corresponding vector fields (26) are determined from the preprocessed images (23), and in a subsequent step (S3), material flow velocity profiles (28) of the crop flow (22) are derived and evaluated from the corresponding vector fields (26).