Multi-Spectral Aerial Imagery Analysis via Blob Partitioning

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current aerial remote sensing technologies face challenges in providing timely and actionable analysis for farmers and organizations, as existing methods require offline processing and are not user-friendly, leading to delayed detection of issues like pest infestations, which can result in irreversible damage due to their reliance on high-resolution imagery that is resource-intensive and impractical for large areas.

Innovation Solution

A method and system for automatic multi-spectral image analysis using blob partitioning, allowing for real-time or near-real-time analysis of large-scale images to identify irregularities, with the option for a second-stage high-resolution analysis of specific areas of interest, guided by GPS data, enabling efficient resource allocation and rapid decision-making.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If high-resolution imagery is used to detect pests and diseases, then detection precision is improved, but resource consumption increases and productivity decreases

Engineering Contradiction:
Improvedetection precisionVSAvoidproductivity
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent divides the image analysis process into two stages: first, low-resolution screening of the entire field to identify suspicious areas; second, high-resolution analysis only of those specific areas. This segmentation of the analysis process enables precise pest detection while avoiding the resource-intensive task of analyzing entire fields at high resolution, thus resolving the contradiction between detection precision and productivity.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If high-resolution imagery is used to detect pests and diseases, then detection precision is improved, but resource consumption increases

Engineering Contradiction:
Improvedetection precisionVSAvoidresource consumption
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent segments the image processing task by first analyzing low-resolution images to identify areas of interest, then applying high-resolution analysis only to those specific segments. This approach maintains detection precision for pests and diseases while significantly reducing the total computational resources required compared to processing entire fields at high resolution.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies high-resolution analysis only partially - specifically to suspicious areas identified in the first stage - rather than excessively analyzing the entire field at high resolution. This partial action approach achieves the necessary detection precision while minimizing resource consumption.

Inventive Principle:
Principle #16Partial or excessive action

3Measurement precision

If offline analysis is used for remote sensing data, then analysis thoroughness is improved, but response time deteriorates

Engineering Contradiction:
Improveanalysis thoroughnessVSAvoidresponse time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent segments the analysis process into immediate automated processing of low-resolution data (providing rapid response) followed by targeted high-resolution analysis only when anomalies are detected (providing thorough analysis when needed). This segmentation enables both fast response time and analysis thoroughness at appropriate stages.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary automated analysis of low-resolution images immediately after acquisition to quickly identify suspicious areas. This preliminary action provides rapid initial response while setting the stage for more thorough high-resolution analysis only when necessary, thus reducing overall response time without sacrificing analysis quality.

Inventive Principle:
Principle #10Preliminary action

4Ease of operation

If automated analysis is implemented, then ease of operation is improved, but analysis accuracy may worsen compared to expert analysis

Engineering Contradiction:
Improveease of operationVSAvoidanalysis accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent segments the analysis workflow so that automated systems handle the initial screening and identification of suspicious areas (improving ease of operation), while preserving the option for expert analysis or more sophisticated automated algorithms to handle the detailed examination of identified anomalies (maintaining analysis accuracy). This segmentation allows non-experts to operate the system easily while still achieving accurate results through targeted detailed analysis.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3374928B1A method for aerial imagery acquisition and analysis
Publication Date: 2024.03.27 AGROWING LTD
  • EP3374928B1 patent drawingFigure 1a
  • EP3374928B1 patent drawingFigure 1b
  • EP3374928B1 patent drawingFigure 1c

AI summary

A method and system for multi-spectral imagery acquisition and analysis, the method including capturing preliminary multi-spectral aerial images according to pre- defined survey parameters at a pre-selected resolution, automatically performing preliminary analysis on site or location in the field using large scale blob partitioning of the captured images in real or near real time, detecting irregularities within the pre- defined survey parameters and providing an output corresponding thereto, and determining, from the preliminary analysis output, whether to perform a second stage of image acquisition and analysis at a higher resolution than the pre-selected resolution. The invention also includes a method for analysis and object identification including analyzing high resolution multi-spectral images according to pre-defined object parameters, when parameters within the pre-defined object parameters are found, performing blob partitioning on the images containing such parameters to identify blobs, and comparing objects confined to those blobs to pre-defined reference parameters to identify objects having the pre-defined object parameters.