Image processing device, image processing method, and program

The method effectively addresses the challenges of existing technologies by generating a reference image based on the first derivative and a marker image using the watershed method to extract contour lines from densely packed images with small brightness differences between objects or images with uneven object sizes, thus enabling accurate extraction of contour lines in images.

WO2026134305A1PCT designated stage Publication Date: 2026-06-25RESONAC CORP
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
RESONAC CORP
Filing Date
2025-12-18
Publication Date
2026-06-25

AI Technical Summary

Technical Problem

Existing image processing methods for extracting contour lines are prone to errors due to noise interference, especially when dealing with images of densely packed objects with small brightness differences or uneven sizes.

Method used

The method involves generating a reference image based on the first derivative and a marker image based on the second derivative of the target image, using Sobel and Laplacian filters respectively, and applying the watershed method to extract contour lines using these images, which are combined to divide the target image into regions, using the watershed method to extract contour lines, thus enabling accurate extraction of contour lines in densely packed images.

Benefits of technology

This method allows for contour lines to be accurately extracted from densely packed images with small differences in brightness between objects or images with uneven sizes, thus enabling efficient creation of training images for a machine learning model that predicts contour lines.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure JP2025044411_25062026_PF_FP_ABST
    Figure JP2025044411_25062026_PF_FP_ABST
Patent Text Reader

Abstract

This image processing device comprises: an image acquiring unit that acquires a target image in which a plurality of objects have been captured; a first image generating unit that generates a first image on the basis of a first-order differentiation of the target image; a second image generating unit that generates a second image on the basis of a second-order differentiation of the target image; and a contour line extracting unit that, on the basis of the first image and the second image, extracts contour lines of the objects captured in the target image.
Need to check novelty before this filing date? Find Prior Art