Stereo Crop Imaging for Automated Yield Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manually monitoring crop health and yield in large agricultural fields is time-consuming, costly, and prone to quality risks due to human error.
Innovation Solution
A plant analysis system with a vehicle-mounted imaging device that generates stereo image data, using machine learning to autonomously detect objects of interest and estimate yield, including a back-end computer system for image processing and yield estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual inspection is used to monitor crop health and yield, then quality control can be maintained, but time consumption and cost increase significantly
Solution Approach 1:
The patent replaces manual mechanical inspection with an automated imaging system that uses cameras and machine learning algorithms to detect and characterize crops. The system captures images, processes them through neural networks, and automatically determines crop health metrics, eliminating the need for human inspectors while maintaining or improving measurement precision.
Solution Approach 2:
The system enables self-service monitoring where the crop field itself provides the information needed for assessment. The imaging system captures data directly from the crops, and the machine learning model automatically analyzes this data to determine health status, yield predictions, and other metrics without requiring external human intervention.
2Productivity
If farmhands are hired to monitor crops, then monitoring can be performed, but quality risk increases due to human error
Solution Approach 1:
The patent replaces human farmhands with an automated machine vision system that uses consistent imaging parameters and machine learning algorithms. This substitution eliminates variability introduced by different inspectors' skills, attention levels, and interpretation methods, ensuring reliable and consistent data quality across all monitoring activities.
Solution Approach 2:
The system standardizes monitoring parameters through fixed imaging settings, controlled lighting conditions, and consistent algorithmic analysis. By changing from variable human parameters to fixed machine parameters, the system achieves reliable and repeatable measurements across different times and locations in the field.
3Area of stationary object
If large fields are monitored manually, then comprehensive coverage can be achieved, but cost and time requirements become prohibitive
Solution Approach 1:
The patent employs a universal imaging system that can monitor entire fields using standardized cameras and processing algorithms. The same hardware and software platform used for small plots can scale to large agricultural fields, providing comprehensive coverage without proportionally increasing system complexity or operational burden.
Data Source
AI summary
Plant analysis system includes a vehicle configured to traverse a field in which the plant is growing and an imaging device mechanically coupled to the vehicle. Imaging device is configured to generate stereo image data associated with the plant. A back-end computer system configured to store a machine learning algorithm that, when executed by a processor, cause the back-end computer system to receive the stereo image data from the imaging device, autonomously detect an object of interest associated with the plant based on the received stereo image data, characterize the detected object of interest, and estimate a crop yield based on the characterization of the detected object of interest.