Object Mapping via Image Segmentation for Precision Spraying
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing agricultural spraying technologies are limited by speed and lighting conditions, and geoprocessing tools struggle to accurately account for extraneous objects in fields, leading to inefficient and time-consuming chemical application processes.
Innovation Solution
A method involving image segmentation, region partitioning, and bounding region formation to accurately map and categorize objects, such as plants, using machine learning and image data processing to enhance precision and efficiency in chemical application.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If camera-mounted sprayers are used to detect targets, then selective chemical application is enabled, but the effectiveness and accuracy are limited by the speed of movement and lighting conditions
Solution Approach 1:
The patent replaces the mechanical camera-based detection system with an optical sensor system that captures images of the field. These images are then processed using image segmentation algorithms to identify plants and generate spray maps, substituting mechanical detection with optical and computational methods to improve reliability while maintaining selective application capability
Solution Approach 2:
The patent introduces an intermediary processing system that includes image capture devices, image segmentation processors, and spray map generators. This intermediary system processes field images to create detailed plant location maps, which then guide the sprayer system, thereby improving accuracy by decoupling detection from direct spraying control
2Productivity
If raster analysis with grid subdivision is used to analyse fields, then targeted chemical application can be guided, but extraneous objects cannot be easily excluded from the analysis
Solution Approach 1:
The patent applies image segmentation to divide the field image into distinct regions, automatically identifying and separating plants from extraneous objects like roads and railway tracks. This segmentation process creates accurate plant location maps that exclude non-target objects, improving measurement precision while maintaining productivity
Solution Approach 2:
The patent changes the analysis approach from fixed grid-based raster analysis to dynamic image segmentation with adjustable parameters. The system can modify segmentation thresholds and parameters to adapt to different field conditions and object types, improving the ability to distinguish plants from extraneous objects while maintaining efficient targeted application
3Loss of information
If image data is processed using traditional raster analysis, then field analysis can be performed, but the process is time-consuming and computationally intensive
Solution Approach 1:
The patent extracts only the relevant information needed for targeted spraying by using image segmentation to directly identify plant locations and generate spray maps. This extraction approach eliminates the need to process entire grid sections, reducing computational intensity and processing time while preserving all necessary field analysis capabilities
Solution Approach 2:
The patent applies partial action by focusing computational resources only on regions containing plants identified through image segmentation, rather than analyzing the entire field grid. This selective processing reduces time and computational resources while maintaining complete field analysis capability for targeted application
Data Source
AI summary
Disclosed herein are methods and systems for mapping objects. A computer-implemented method for mapping objects comprises receiving image data of an area, and applying an image segmentation process to the image data to identify one or more elements of the image data determined to have a likelihood of representing one or more predetermined objects in the area. The method further comprises assigning a value to each element of the image data based on the determined likelihood of the element representing the one or more predetermined objects, partitioning the image data into two or more overlapping layers of regions, determining a quality of fit of one or more regions of each layer of the two or more layers to each of the one or more predetermined objects using the values assigned to the elements of the image data, and forming one or more bounding regions from one or more regions of one or more layers of the two or more layers based on the quality of fit of the respective one or more regions. Each bounding region encloses one or more of the one or more predetermined objects. The method further comprises outputting region data representing the one or more bounding regions. Also disclosed here are methods and systems for detecting changes in objects.