Waste Object Identification Using Imaging and ML Separation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Bulky or slow-burning objects in solid waste cause inefficiencies and increased costs in waste-to-energy systems by clogging incinerator hoppers and requiring supplementary fuel, as they are difficult to identify and remove manually.

Innovation Solution

Implementing imaging devices and a machine learning model to identify target objects within solid waste, providing visual representations and automated removal instructions to separators.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual monitoring and removal by crane operator is used, then operational flexibility is maintained, but identification accuracy and processing efficiency deteriorate due to difficulty in identifying target objects within waste volume

Engineering Contradiction:
Improvetarget object identification accuracyVSAvoidwaste processing efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces the manual mechanical inspection method with an automated imaging and machine learning system. Imaging devices capture images of waste material, and a machine learning model automatically identifies target objects, substituting human visual inspection with automated optical and computational systems to improve both accuracy and efficiency

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

Solution Approach 2:

The patent introduces an intermediary system consisting of imaging devices and machine learning algorithms between the waste material and the crane operator. This intermediary automatically processes and analyzes waste images to identify target objects, providing enhanced detection capabilities that neither manual inspection alone nor automated removal alone could achieve

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If automated separator is implemented, then processing speed increases, but system complexity and cost increase due to need for advanced detection and control systems

Engineering Contradiction:
Improvewaste processing speedVSAvoidseparator system complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent segments the waste processing system into distinct functional modules: imaging devices for capture, machine learning model for analysis, and automated separator for physical removal. This segmentation allows each component to be optimized independently and simplifies the overall system architecture by dividing complex functions into manageable parts

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The machine learning model performs self-learning and automatic identification of target objects without requiring constant human intervention or complex control algorithms. The system trains on waste images and autonomously identifies target objects, reducing the complexity of the control system while maintaining high processing speed

Inventive Principle:
Principle #25Self-service

3Reliability

If target objects are not removed, then processing continuity is maintained, but harmful effects increase due to clogging and supplementary fuel requirements

Engineering Contradiction:
Improveincinerator operation continuityVSAvoidclogging and fuel consumption
Core Design Contradiction:
ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The patent implements preliminary identification and removal of target objects before they can cause clogging or combustion problems in the incinerator. The imaging and machine learning system detects target objects in advance, and the automated separator removes them proactively, preventing harmful effects rather than reacting to them after they occur

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent converts the potential harm of target objects (which would cause clogging and require supplementary fuel) into a benefit by using the imaging and machine learning system to identify and separate them. The very objects that could cause problems become the focus of an automated detection and removal system that improves overall incinerator efficiency and reliability

Inventive Principle:
Principle #22Blessing in disguise (Convert harm into benefit)

Data Source

PatentEP4002295B1Target object identification for waste processing
Publication Date: 2026.03.25 ACCENTURE GLOBAL SOLUTIONS LTD
  • EP4002295B1 patent drawingFigure 1A
  • EP4002295B1 patent drawingFigure 1B
  • EP4002295B1 patent drawingFigure 2

AI summary

In some implementations, a target object identification system may train a machine learning model to identify a target object in a volume of solid waste. The target object identification system may receive, from at least one imaging device positioned in a waste processing facility, imaging data associated with a portion of a volume of solid waste. The target object identification system may identify, using the trained machine learning model and based at least in part on the imaging data, the target object, wherein the target object is disposed within the volume of solid waste. The target object identification system may provide, to a separator control assembly that controls a separator, an output associated with the target object, wherein the output is to facilitate removal of the target object from the volume of solid waste.