Method, non-transitory computer-readable storage medium, and apparatus for searching image database
Neural networks are used to remove noise from patent images, enhancing the accuracy and efficiency of image retrieval by isolating underlying objects, thus improving the search process in large databases.
JP2025114765APending Publication Date: 2025-08-05CAMELOT UK BIDCO LTD
View PDF 0 Cites 0 Cited by
Patent Information
- Application Number
- JP2025079546
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2021-05-28
- Filing Date
- 2025-05-12
- Publication Date
- 2025-08-05
AI Technical Summary
Technical Problem
Existing image matching algorithms are hindered by image noise such as annotations and labels in patent images, leading to inaccurate and inefficient searches.
Method used
A method involving neural networks to remove noise from images, including a first neural network for annotation removal and a second neural network for contour masking, followed by correlation with a reference database to select matching images based on threshold correlation values.
Benefits of technology
Enables accurate and efficient image retrieval by removing noise from patent drawings, facilitating effective comparison and identification of similar images in large databases.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure 2025114765000001_ABST
Abstract
To provide a method, a non-transitory computer-readable storage medium, and an apparatus for searching an image database.SOLUTION: A method for searching an image database including applying a masking to an adulterated image comprises: receiving an adulterated image of an object, the adulterated image including object annotations for visual reference; applying a first neural network to the adulterated image; associating a result of applying the first neural network with each image of a reference database of images, the result including an edited image of the object and each image of the reference database of images including a reference object; and selecting, as a matching image, one or more images of the reference database of images having correlation values above a threshold correlation value.SELECTED DRAWING: Figure 2A
Need to check novelty before this filing date? Find Prior Art