Acoustic Pre-filtering for Image Recognition Speed
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image-based produce identification systems in retail environments face long recognition times due to the need to compare captured produce image data against a large database of items, leading to inefficient identification processes.
Innovation Solution
Incorporating an acoustic energy source and data collector to emit and capture acoustic energy from produce items, which processes and compares the deflected energy to reduce the number of potential matches, thereby narrowing down the database search and accelerating image recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image recognition is used to identify produce items by comparing captured image data against a large database, then identification accuracy is maintained, but recognition time becomes excessively long
Solution Approach 1:
The patent segments the identification process into two stages: first using acoustic data to narrow down the database to a small subset of candidate items, then using image recognition to identify the specific item from this reduced set. This segmentation dramatically reduces the search space for image recognition while maintaining accuracy.
Solution Approach 2:
The patent performs preliminary acoustic data collection and processing before image recognition. The acoustic data is used to pre-filter and categorize items, creating a narrowed-down list of candidates that the image recognition system then processes. This preliminary action reduces the computational burden and time required for the main identification task.
2Measurement precision
If a large database of items is used for comprehensive product identification, then coverage and accuracy are improved, but processing time and system complexity increase
Solution Approach 1:
The patent divides the large database into categories or subsets based on acoustic characteristics. Instead of processing the entire large database with image recognition, the system first uses acoustic data to segment the database into smaller, more manageable subsets, then processes only these subsets with image recognition.
Solution Approach 2:
The patent introduces acoustic data as an intermediary between the large database and the image recognition process. The acoustic data serves as a mediator that filters and pre-sorts the database, creating a bridge that reduces the complexity of directly processing the entire large database with image recognition.
3Loss of time
If acoustic data collection is added to the recognition system, then recognition time is reduced by narrowing down candidates, but device complexity increases
Solution Approach 1:
The patent combines acoustic data collection and image recognition into a unified identification system. The acoustic module and image recognition module work together in an integrated manner, where the output of the acoustic module feeds into the image recognition module, creating a synergistic system that reduces overall processing time.
Solution Approach 2:
The patent creates a multi-functional identification system that can process both acoustic data and image data. The system is designed to handle multiple types of input data (acoustic and visual) and uses both modalities to achieve fast and accurate identification, making the system universally applicable to various produce items.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach significantly reduces recognition times by categorizing items into broad classes, such as rigid, semi-rigid, or soft, thereby limiting the search to a smaller subset of items, potentially reducing recognition times by 50 times or more.
Implementation Method 1
an acoustic energy source arranged to emit acoustic energy towards the product item. The product recognition further comprises a product acoustic data collector arranged to (i) capture acoustic energy deflected from the product item
Data Source
AI summary
A product recognition system comprises a product image data collector arranged to capture image data which is representative of an image of a product item. The product recognition system also comprises an acoustic energy source arranged to emit acoustic energy towards the product item. The product recognition further comprises a product acoustic data collector arranged to (i) capture acoustic energy deflected from the product item, (ii) process the captured acoustic data which has been deflected from the product item to provide product acoustic data which is representative of one or more characteristics of the product item, and (iii) compare the product acoustic data with a store of reference acoustic data to provide one or more subsets of items against which the captured image data can be compared to identify the product item.