Plant Treatment Model Selection for Precision Weed Control

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for applying plant treatments in agricultural fields are labor-intensive and costly when treating individual plants, as they require manual intervention and lack flexibility to accommodate varying user preferences and changing conditions.

Innovation Solution

A system that utilizes plant treatment models trained based on user preferences, allowing for customized treatment plans by providing images of plants for user interaction and annotation, and configuring farming machines to perform targeted treatments based on these preferences.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If manual treatment application is used for individual plants, then treatment precision is improved, but labor intensity and cost increase

Engineering Contradiction:
Improvetreatment precisionVSAvoidlabor efficiency
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

The system enables automated self-service treatment by using image recognition models to automatically identify plants requiring treatment and control farming machines to apply treatments autonomously, eliminating manual labor while maintaining precision

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

Manual mechanical treatment application is replaced with an automated system combining image capture devices, machine learning models, and controlled farming machines that use optical and computational methods to identify and treat plants

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Productivity

If uniform treatment is applied to large areas, then treatment efficiency is improved, but treatment precision and customization are lost

Engineering Contradiction:
Improvetreatment efficiencyVSAvoidtreatment precision
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The system applies different treatment actions to different local areas based on individual plant characteristics. Each plant is analyzed separately through image recognition, and treatment decisions are customized locally rather than applying uniform treatment across the entire field

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The field is segmented into individual plant units for separate analysis and treatment. The image recognition system processes each plant independently, identifying specific plants that require treatment while leaving others untreated, enabling precise localized intervention

Inventive Principle:
Principle #1Segmentation

3Measurement precision

If multiple plant treatment models are trained and evaluated, then model selection accuracy is improved, but system complexity increases

Engineering Contradiction:
Improvemodel selection accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system uses user interactions with displayed plant images as feedback to evaluate and select the most appropriate treatment model. User preferences and corrections are fed back into the system to refine model selection, ensuring accuracy while managing complexity through iterative improvement

Inventive Principle:
Principle #23Feedback

4Reliability

If customized treatment models are created for each user, then treatment effectiveness is improved, but training time and computational resources increase

Engineering Contradiction:
Improvetreatment effectivenessVSAvoidmodel training time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

Multiple plant treatment models are pre-trained in advance on diverse plant data before deployment. When deployed, the system selects from these pre-trained models based on current field conditions and user preferences, avoiding the need to train custom models from scratch for each situation and reducing training time

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS12628810B2Plant treatment model selection based on agricultural image interaction
Publication Date: 2026.05.19 DEERE & CO
  • US12628810B2 patent drawing
  • US12628810B2 patent drawing
  • US12628810B2 patent drawing

AI summary

Embodiments relate to selecting and utilizing a plant treatment model. A control system may provide images of plants of different dimensions and types for display to a user. The control system may generate one or more plant treatment action preferences of the user based on user interactions with the provided images of plants. The control system may apply trained plant treatment models to the images of plants, each plant treatment model identifying one or more plants in the images and one or more corresponding plant treatment actions. The control system may select a plant treatment model based on a comparison of (1) plants and corresponding plant treatment actions identified by the plant treatment models in the images and (2) the one or more plant treatment action preferences of the user. The control system may configure a farming machine to operate based on the selected plant treatment model.