Robot Lawn Mower Obstacle Recognition Using Hue Histograms
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional weeding robots require manual calibration of weeding region boundaries using boundary lines, consuming manpower and material resources, and are limited in shape due to 90-degree corner restrictions, affecting recognition efficiency and accuracy.
Innovation Solution
An obstacle recognition method and apparatus that utilizes hue information to generate a target hue histogram, identify peak information and hue range, and determine obstacles in weeding regions, eliminating the need for manual boundary calibration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If boundary lines are buried to calibrate weeding regions, then the weeding region boundaries can be defined, but manpower and material resources are consumed and costs increase
Solution Approach 1:
The patent extracts the boundary calibration process from physical boundary lines and transfers it to virtual boundary definition through image processing. The hue histogram method extracts color information to automatically determine weeding region boundaries without requiring physical markers, thereby eliminating the consumption of materials and manpower associated with burying boundary lines.
Solution Approach 2:
The patent replaces the mechanical system of burying and physically marking boundary lines with an optical and computational system. By using hue histogram analysis of images captured by sensors, the system automatically identifies weeding region boundaries through color differentiation, substituting mechanical boundary marking with optical detection and algorithmic processing.
2Adaptability or versatility
If boundary lines are buried for region calibration, then weeding regions can be defined, but the shape is limited due to 90-degree corner restrictions
Solution Approach 1:
The patent introduces dynamic boundary definition through image processing rather than static physical boundary lines. The hue histogram method allows boundaries to adapt to any shape by analyzing color distributions in captured images, enabling the system to handle arbitrary geometries including curved lines and irregular shapes that cannot be achieved with rigid 90-degree corner constraints.
Solution Approach 2:
The patent changes the parameter used for boundary definition from physical geometry (90-degree corners) to color space parameters (hue values). By transforming the boundary calibration problem into a color analysis problem using hue histograms, the system can define regions with any shape without being constrained by angular restrictions, thus improving adaptability while reducing calibration complexity.
3Measurement precision
If conventional obstacle recognition methods are used, then obstacles can be detected, but recognition efficiency and accuracy are insufficient
Solution Approach 1:
The patent utilizes color changes and color distribution analysis through hue histograms to improve obstacle recognition. By analyzing the hue information and its distribution in the image, the system can distinguish obstacles from the background based on color characteristics, significantly improving recognition accuracy and efficiency compared to conventional methods that do not leverage color space analysis.
Solution Approach 2:
The patent introduces hue histogram analysis as an intermediary step between image capture and obstacle detection. This intermediary processing extracts color distribution characteristics that serve as powerful features for obstacle identification, enabling more accurate and efficient recognition by providing additional discriminative information beyond simple pixel intensity analysis.
Data Source
AI summary
An obstacle recognition method includes the steps of obtaining hue information of a candidate weeding region image; generating a target hue histogram of the candidate weeding region image according to the hue information, and obtaining peak information and hue range information of the target hue histogram; and determining whether there are obstacles in the candidate weeding region image according to the peak information and the hue range information. Related apparatus, electronic devices, computer readable storage media, and weeding robots are disclosed.