Image processing device and image processing method
An image processing device ( 10 ) according to an aspect of the present invention includes an acquiring section ( 151 ) and an estimation section ( 152 ). The acquiring section ( 151 ) acquires a first image and a second image obtained by capturing randomly stacked workpieces (W 1 a , W 1 b ). The estimation section ( 152 ) generates a matching map of a feature of the first image and a feature of the second image, estimates a position, an orientation and a classification score of each workpiece as a target for each of the first image and the second image, and estimates the position of the workpiece, based on a matching result using the matching map and an estimation result of the position.
1. An image processing device comprising:
an acquiring section configured to acquire a first image and a second image obtained by capturing randomly stacked workpieces; and
an estimation section configured to generate a matching map of a feature of the first image and a feature of the second image, estimate a position, an orientation and a classification score of each workpiece as a target for each of the first image and the second image, and estimate the position of the workpiece, based on a matching result using the matching map and an estimation result of the position.
2. The image processing device according to claim 1 , wherein
the acquiring section is a stereo camera; and
the estimation section calculates a depth of the workpiece from the stereo camera.
3. The image processing device according to claim 1 , wherein the estimation section further detects an obstacle other than the workpiece in at least one of the first image and the second image.
4. An image processing method comprising:
by a computer, acquiring a first image and a second image obtained by capturing randomly stacked workpieces; and
by the computer, generating a matching map of a feature of the first image and a feature of the second image, estimating a position, an orientation and a classification score of each workpiece as a target for each of the first image and the second image, and estimating the position of the workpiece, based on a matching result using the matching map and an estimation result of the position.