Object Identification Using Distance and Luminance Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing object identification techniques, such as those described in JP-A-2010-191745, often produce incorrect results when multiple objects are imaged together, leading to unreliable identification outcomes.
Innovation Solution
The method involves acquiring a distance image and a luminance image of an object using separate cameras, where the distance image represents the distance of objects from a camera position and the luminance image represents the luminance of reflected light. An indicator value is calculated based on the distance image to assess the reliability of the identification result, and if below a threshold, the identification is invalidated.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If template matching is used to identify object types in captured images, then identification can be performed using simple image comparison, but wrong identification results occur when another object is imaged over a part of the identification target object
Solution Approach 1:
The patent introduces a depth dimension by acquiring depth information for each pixel in the captured image. This allows the system to distinguish between objects at different distances from the camera, resolving ambiguities in 2D template matching when objects overlap in the image plane. The depth map enables the system to determine which object is in front and which is in the background, thereby improving identification accuracy.
2Device complexity
If only luminance information from a single camera is used for identification, then the system remains simple, but reliability decreases when multiple objects are present in the image
Solution Approach 1:
The patent segments the imaging information into two distinct components: luminance information from a color camera and depth information from a depth camera. This segmentation allows each sensor to capture specific types of data optimally, with the depth camera providing distance information that helps separate overlapping objects in the scene, thereby improving identification reliability without requiring a single complex sensor system.
Data Source
AI summary
There is provided an identification method acquiring a first image, a pixel value of each of pixels of which represents a distance from a first position to an imaging target object including an identification target object, acquiring a second image captured from the first position or a second position different the first position, a pixel value of each of pixels of the second image representing at least luminance of reflected light from the imaging target object, identifying a type of the identification target object based on the second image, and calculating, based on the first image, an indicator value indicating a reliability degree of an identification result of the type of the identification target object based on the second image.