3-dimensional object recognition method and bin-picking system using the method

A 3-dimensional object recognition method comprising:

• a step (A) of taking a pair of first and second images for making a stereo image of an object, and a third image from a viewpoint different from them;
• a step (B) of detecting a two-dimensional feature of said object in each of the first, second and third images;
• a step (D) of making a correspondence of the two-dimensional feature each between the first and second images and between the first and third images according to a stereoscopic measurement principle;
• a step (E) of evaluating a degree of reliability of a result of the step (D) by comparing the two-dimensional feature detected in the first image with the corresponding two-dimensional features detected in the second and third images;
• a step (F) of recognizing a first candidate of three-dimensional position and posture of said object according to information in three dimensions of the two-dimensional feature obtained from the correspondence between the first and second images, and a second candidate of three-dimensional position and posture of said object according to information in three dimensions of the two-dimensional feature obtained from the correspondence between the first and third images;
• a step (G) of evaluating a degree of reliability of each of the first and second candidates;
• and
• a step (H) of determining three-dimensional position and posture of said object according to the first and second candidates provided from the step (G).

A 3-dimensional object recognition method comprising:

• a step (A) of taking a pair of first and second images for making a stereo image of an object;
• a step (B) of detecting a two-dimensional feature of said object in each of the first and second images;
• a step (D) of making a correspondence of said two-dimensional feature between the first and second images according to a stereoscopic measurement principle;
• a step (F) of recognizing three-dimensional position and posture of said object according to information in three dimensions of said two-dimensional feature obtained by the correspondence;
• and
• a step (G) of evaluating a degree of reliability of the recognized three-dimensional position and posture,
• wherein the method comprises at least one of a step (C) of evaluating a degree of reliability of a result of the step (B) by comparing with known model data of said object, which is performed between the steps (B) and (D), and a step (E) of evaluating a degree of reliability of a result of the step (D) by comparing said two-dimensional feature detected in the first image with a corresponding two-dimensional feature detected in the second image, which is performed between the steps (D) and (F), and
• wherein the step (G) includes comparing the recognized three-dimensional position and posture of said object in the step (F) with a known three-dimensional feature of said object, and giving a rank in a degree of reliability to the recognized three-dimensional position and posture according to a difference therebetween.

A 3-dimensional object recognition method comprising:

• a step (A) of taking a pair of first and second images for making a stereo image of an object;
• a step (B) of detecting a two-dimensional feature of said object in each of the first and second images;
• a step (D) of making a correspondence of said two-dimensional feature between the first and second images according to a stereoscopic measurement principle;
• a step (F) of recognizing three-dimensional position and posture of said object according to information in three dimensions of said two-dimensional feature obtained by the correspondence;
• a step (G) of evaluating a degree of reliability of the recognized three-dimensional position and posture;
• a step (H) of predicting a two-dimensional feature on each of the first and second images according to the three-dimensional position and posture obtained by the step (F);
• and
• a step (I) of evaluating a degree of agreement between the predicted two-dimensional feature and said two-dimensional feature detected in the step (B),
• wherein the method comprises at least one of a step (C) of evaluating a degree of reliability of a result of the step (B) by comparing with known model data of said object, which is performed between the steps (B) and (D), and a step (E) of evaluating a degree of reliability of a result of the step (D) by comparing said two-dimensional feature detected in the first image with a corresponding two-dimensional feature detected in the second image, which is performed between the steps (D) and (F).

A bin-picking system comprising:

• an image-processing unit for performing a 3-dimensional object recognition method comprising:
• a step (A) of taking a pair of first and second images for making a stereo image of an object;
• a step (B) of detecting a two-dimensional feature of said object in each of the first and second images;
• a step (D) of making a correspondence of said two-dimensional feature between the first and second images according to a stereoscopic measurement principle;
• a step (F) of recognizing three-dimensional position and posture of said object according to information in three dimensions of said two-dimensional feature obtained by the correspondence;
• and
• a step (G) of evaluating a degree of reliability of the recognized three-dimensional position and posture,
• wherein the method comprises at least one of a step (C) of evaluating a degree of reliability of a result of the step (B) by comparing with known model data of said object, which is performed between the steps (B) and (D), and a step (E) of evaluating a degree of reliability of a result of the step (D) by comparing said two-dimensional feature detected in the first image with a corresponding two-dimensional feature detected in the second image, which is performed between the steps (D) and (F);
• a robot with a flexible robot arm;
• and
• a robot controller for controlling said robot according to information provided from said image-processing unit such that said robot arm picks up an object from a bin, in which a plurality of objects are heaped up in confusion, and carries the picked-up object to a desired position.

A bin-picking system comprising:

• an image-processing unit for performing a 3-dimensional object recognition method comprising:
• a step (A) of taking a pair of first and second images for making a stereo image of an object, and a third image from a viewpoint different from them;
• a step (B) of detecting a two-dimensional feature of said object in each of the first, second and third images;
• a step (D) of making a correspondence of the two-dimensional feature each between the first and second images and between the first and third images according to a stereoscopic measurement principle;
• a step (F) of recognizing a first candidate of three-dimensional position and posture of said object according to information in three dimensions of the two-dimensional feature obtained from the correspondence between the first and second images, and a second candidate of three-dimensional position and posture of said object according to information in three dimensions of the two-dimensional feature obtained from the correspondence between the first and third images;
• a step (G) of evaluating a degree of reliability of each of the first and second candidates;
• wherein the recognition method comprises at least one of a step (C) of evaluating a degree of reliability of a result of the step (B) by comparing with a model data of said object, which is performed between the steps (B) and (D), and a step (E) of evaluating a degree of reliability of a result of the step (D) by comparing the two-dimensional feature detected in the first image with the corresponding two-dimensional features detected in the second and third images, which is performed between the steps (D) and (F), and
• a step (H) of determining three-dimensional position and posture of said object according to the first and second candidates provided from the step (G);
• a robot with a flexible robot arm;
• and
• a robot controller for controlling said robot according to information provided from said image-processing unit such that said robot arm picks up an object from a bin, in which a plurality of objects are heaped up in confusion, and carries the picked-up object to a desired position.