Press Brake Gaze Image Extraction Using Distance-Based Object Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The complexity of the press brake structure and similar overall colors make it difficult for users to efficiently gaze at the desired object within captured images, as various other objects are present and hard to distinguish.

Innovation Solution

A press brake system incorporating an image capturing device, a distance measuring device, and an image processing device that captures images of the work region, detects distances to objects, and generates a gaze image by cutting out the desired gaze object from the captured image based on distance data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Area of stationary object

If a plurality of image capturing devices are used to capture the work region, then the coverage of the work region is improved, but the complexity of the system increases and the captured image contains various other objects that make it difficult to identify the gaze object

Engineering Contradiction:
Improvecoverage of work regionVSAvoidsystem complexity
Core Design Contradiction:
Area of stationary objectVSDevice complexity

Solution Approach 1:

The patent extracts the gaze object from the complex captured image by using depth information from a distance measuring device. The image processing device separates the gaze object based on distance data, removing background and other irrelevant objects to create a simplified view that highlights only the target object of interest.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent introduces a distance measuring device as an intermediary that provides depth information to differentiate the gaze object from other objects in the captured image. This intermediary device enables the image processing device to segment and isolate the gaze object based on distance criteria.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Loss of information

If the captured image displays all objects in the work region, then complete information is provided, but the user cannot efficiently gaze at the desired object due to similar colors and multiple objects present

Engineering Contradiction:
Improveinformation completenessVSAvoidvisual recognition efficiency
Core Design Contradiction:
Loss of informationVSEase of operation

Solution Approach 1:

The image processing device extracts only the gaze object from the complete captured image by utilizing distance information. This extraction process maintains the essential visual information of the gaze object while removing distracting background elements and other objects, thereby improving visual recognition efficiency without significant loss of relevant information.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies local quality enhancement by focusing visual attention on the gaze object through selective display. The processed image emphasizes the local region containing the gaze object while suppressing or removing other regions, creating a visual hierarchy that guides user attention to the most important area.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP4215292B1Press brake and image output method
Publication Date: 2025.05.21 AMADA CO LTD
  • EP4215292B1 patent drawingFigure 1
  • EP4215292B1 patent drawingFigure 2
  • EP4215292B1 patent drawingFigure 3

AI summary

A camera (50) captures an image of a work region in which an upper table (26) and a lower table (22) carry out a bending process in a press brake main body (10), and outputs a captured image (ID). A distance measuring device (55) detects a distance to an object present in the captured image (ID) and generates distance data (DD) in which the object and the distance are associated with each other. An image processing device (60) generates, based on the distance data (DD), a gaze image (CD) obtained by cutting out, from the captured image (ID), a gaze object that is an object to be gazed at within the work region, and outputs the gaze image (CD) to a target device usable by a user.