Agricultural Row Detection Using 3D Ridge Height Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing row detection systems in agricultural machines are prone to accuracy degradation due to disturbance factors such as varying daylighting conditions and crop growth states.
Innovation Solution
A row detection system utilizing an imaging device and a processing device to perform feature point matching and perspective projection, estimating ridge heights based on feature point movements to enhance detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional image processing methods are used for row detection, then the system is simple to implement, but detection accuracy degrades under varying daylighting conditions and crop growth states
Solution Approach 1:
The patent transitions from 2D image analysis to 3D spatial analysis by calculating real-world coordinates (X, Y, Z) from image coordinates using camera parameters and perspective projection. This dimensional transformation enables height estimation and ridge detection that is invariant to lighting conditions and crop appearance variations.
Solution Approach 2:
The patent introduces an intermediary coordinate transformation system that maps image plane coordinates to real-world 3D coordinates through perspective projection and camera calibration parameters. This intermediary layer decouples the detection algorithm from direct pixel intensity analysis, making it robust against daylighting variations.
2Reliability
If feature point matching across multiple images is performed, then detection robustness against daylighting variations is improved, but processing time increases
Solution Approach 1:
The patent extracts only the essential geometric features (feature points and their coordinates) from images, discarding intensity and color information that vary with lighting. This extraction of invariant geometric properties maintains robustness while reducing processing complexity compared to full image analysis.
Solution Approach 2:
The patent uses feature point matching to create a correspondence between feature points across multiple images, effectively copying spatial relationships from one image to another. This copying of geometric structure enables temporal averaging and robustness improvement without requiring full image processing.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
A row detection system includes: an imaging device mounted to an agricultural machine, the imaging device being configured to image a ground surface that is traveled by the agricultural machine to acquire time-series images containing at least a portion of the ground surface; and a processing device configured to perform image processing for the time-series images. The processing device is configured to: from a plurality of images among the time-series images that have been acquired at different points in time, determine a first amount of movement of each of the plurality of feature points in an image plane through feature point matching; through perspective projection of each of the plurality of feature points from the image plane onto a reference plane corresponding to the ground surface; determine a second amount of movement of each projection point in the reference plane based on the first amount of movement; and, based on the second amount of movement, estimate heights of the plurality of feature points from the reference plane to detect a ridge on the ground surface.