Liquid Volume Estimation From Transparent Container Images

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Estimating the volume of liquids in translucent or transparent containers is challenging due to varying shapes and blending of the background with the container, making it difficult to determine the volume accurately.

Innovation Solution

A method using a single input image of a transparent or translucent container, processed with an image background removal algorithm and machine-learning algorithms like CNN and Mask R-CNN, to predict the volume of liquid by enhancing contrast and using heatmaps for improved pixel distinction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If image background removal algorithm is applied, then measurement precision is improved, but device complexity increases

Engineering Contradiction:

Improvevolume estimation accuracyVSAvoidprocessing system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary processing pipeline consisting of background removal algorithm and machine learning model that mediates between the input image and volume estimation. This intermediary system processes the image through multiple stages (background removal, feature extraction, volume prediction) to achieve accurate measurements without requiring complex specialized hardware, thus resolving the contradiction between measurement precision and device complexity.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If machine-learning algorithm is used, then measurement precision is improved, but loss of time increases

Engineering Contradiction:

Improvevolume estimation accuracyVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies preliminary action by pre-processing the image through background removal algorithm before feeding it to the machine learning model. This preliminary processing step prepares the image data in advance, extracting relevant features and removing distractions, which enables the machine learning model to make accurate predictions more efficiently, thus reducing overall processing time while maintaining high measurement precision.

Inventive Principle:

Principle #10Preliminary action

3Ease of operation

If single input image is processed, then ease of operation is improved, but measurement precision worsens

Engineering Contradiction:

Improveuser operation simplicityVSAvoidvolume estimation accuracy

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The patent replaces manual measurement methods with an automated image processing system that uses machine learning algorithms. Instead of requiring users to manually measure liquid volume in transparent containers, the system automatically processes a single input image through background removal and neural network inference to estimate volume, thus maintaining ease of operation while significantly improving measurement precision through intelligent automation.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP4581575B1Estimating volumes of liquid

Publication Date: 2025.12.31 KONINKLIJKE PHILIPS NV

Figure 1
Figure 2
Figure 3

AI summary

Proposed concepts thus aim to provide schemes, solutions, concept, designs, methods and systems pertaining to estimating a volume of liquid in an at least partially transparent container. This can be achieved by automatically processing an input image of a transparent/translucent container containing liquid in order to obtain a background-modified image, wherein the background image data has been removed or altered. A prediction of the volume of liquid in the container may then be determined by processing the background-modified image with a machine-learning algorithm.