Image-Based Crop Yield Prediction With Distance-Corrected Area
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for predicting crop harvest amounts suffer from low prediction accuracy.
Innovation Solution
An information processing device that extracts a target object area from images using a distance sensor to correct object size and predicts production amounts based on corrected sizes using a regression model.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If neural network correction is used to predict harvest amount, then prediction capability is provided, but prediction accuracy is not necessarily high
Solution Approach 1:
The patent changes the key parameter from direct neural network prediction of harvest amount to prediction of crop area, which is then converted to harvest amount through a regression model. This parameter transformation (from complex harvest amount directly to measurable crop area) improves prediction accuracy while maintaining reliability.
Solution Approach 2:
The patent replaces the neural network system with a combination of photographing device, distance sensor, and regression model. This substitution uses optical measurement and mathematical conversion instead of complex neural network processing, achieving both accuracy and reliability.
2Measurement precision
If crop area is measured from photographed images without distance correction, then measurement process is simple, but measurement precision is low
Solution Approach 1:
The patent introduces a distance sensor as an intermediary device that provides distance information between the photographing device and the crop. This intermediary measurement enables accurate area calculation by providing the scaling factor needed to convert image pixels to real-world dimensions, thereby improving measurement precision without excessive complexity.
Solution Approach 2:
The patent adds the distance dimension to the two-dimensional image analysis. By incorporating distance information from the distance sensor, the system transforms a purely 2D image measurement problem into a 3D measurement problem, enabling accurate calculation of actual crop area from the photographed images.
3Measurement precision
If direct harvest amount prediction is attempted without area correction, then prediction process is straightforward, but prediction accuracy is low
Solution Approach 1:
The patent performs preliminary correction of the crop area using distance information before predicting the harvest amount. This preliminary action (area correction) ensures that the basis for harvest prediction is accurate, thereby improving overall prediction accuracy while maintaining efficiency through a systematic two-step process.
Solution Approach 2:
The patent replaces direct harvest amount prediction with a two-stage system: first predicting crop area from images, then converting area to harvest amount using a regression model. This substitution breaks down a difficult prediction task into two more manageable and accurate steps, improving overall accuracy without significantly reducing productivity.
Data Source
AI summary
The present invention is to predict a production amount of a target object with higher accuracy.An information processing device that predicts a production amount of a target object, the information processing device including: an area extraction module configured to extract a target object area in which the target object exists in a photographed image of the target object photographed by a photographing device; a correction module configured to correct a size of the target object area on the basis of a distance from the photographing device to the target object, detected by a distance sensor; and a prediction module configured to predict the production amount of the target object on the basis of a size of the target object area corrected by the correction module.