Detecting object presence on a target surface

• US 9,978,135 B2
• Filed: 02/27/2015
• Issued: 05/22/2018
• Est. Priority Date: 02/27/2015
• Status: Active Grant

• Current Assignee: Cognex Corporation
• Original Assignee: Cognex Corporation
• Inventors: Marrion, Cyril C., Mullan, Nickolas James
• Primary Examiner(s): WOLDEMARIAM, AKLILU K
• Application Number:
  US14/634,717
• Publication Number:
  US 20160253793A1
• Time in Patent Office: 1,180 Days
• Field of Search: 382103, 382141, 382154, 345419, 345427, 348 47, 348362, 340555
• CPC Class Codes: G06F 18/00 Pattern recognition
• CPC Class Codes: G06T 2207/10012 Stereo images
• CPC Class Codes: G06T 2207/10028 Range image; Depth image; 3D point clouds
• CPC Class Codes: G06T 2207/30208 Marker matrix
• CPC Class Codes: G06T 7/0008 checking presence/absence
• CPC Class Codes: G06T 7/85 Stereo camera calibration

Abstract

Described are methods, systems, apparatus, and computer program products for determining the presence of an object on a target surface. A machine vision system includes a first image capture device configured to image a first portion of a target surface from a first viewpoint and a second image capture device configured to image a second portion of the target surface from a second viewpoint. The machine vision system is configured to acquire a first image from the first image capture device, a second image from the second image capture device, rectify the first image and second image, retrieve a disparity field, generate difference data by comparing, based on the mappings of the disparity field, image elements in the first rectified image and a second image elements in the second rectified image; and determine whether the difference data is indicative of an object on the target surface.

Claims

1. A machine vision system comprising:

a first image capture device configured to image a first portion of a target surface from a first viewpoint;

a second image capture device configured to image a second portion of the target surface from a second viewpoint, wherein the first portion of the target surface at least partially overlaps the second portion of the target surface;

a machine vision processor connected to the first image capture device and the second image capture device, the machine vision processor configured to;

acquire a first image of the target surface from the first image capture device;

acquire a second image of the target surface from the second image capture device;

rectify the first image to form a first rectified image and rectify the second image to form a second rectified image;

retrieve a disparity field that provides mappings from a first plurality of coordinates in the first rectified image to a second plurality of coordinates in the second rectified image;

generate difference data by applying one or more difference functions to a first plurality of image elements in the first rectified image and a corresponding second plurality of image elements in the second rectified image, wherein a correspondence between an element of the first plurality of images and an element of the second plurality of images is based on the disparity field, and wherein the difference function is a function for determining a difference in pixel values, including at least one of;

a difference in grey scale values, a difference in hue values, a difference in gradient values, or a difference in reflectance values; and

determine a presence or an absence of an object on the target surface based on the difference data by applying a filtering algorithm to the difference data, the filtering algorithm comprising setting the plurality of difference values to zero if the plurality of difference values are below a threshold.

2. The machine vision system of claim 1, wherein the first image capture device and the second image capture device are a camera and, wherein the camera is configured to image the first portion of the target surface from the first viewpoint and image the second portion of the target surface from the second viewpoint approximately simultaneously.

3. The machine vision system of claim 1, wherein the first image capture device is a first camera and the second image capture device is a second camera.

4. The machine vision system of claim 1, wherein the machine vision processor is configured to acquire the first image and the second image substantially simultaneously.

5. The machine vision system of claim 1, wherein the first plurality of image elements is a first plurality of pixel values.

6. The machine vision system of claim 1, wherein the target surface is one of a conveyor belt, a tray or a receptacle.

7. The machine vision system of claim 1, wherein the machine vision processor is configured to apply a shape-finding algorithm to the difference data to determine a presence or an absence of an object on the target surface.

15. The machine vision system of claim 7, wherein the shape-finding algorithm comprises blob analysis.

8. A method executed by a machine vision system comprising:

acquiring, by the machine vision system, a first image of the target surface from a first viewpoint;

acquiring, by the machine vision system, a second image of the target surface from a second viewpoint, wherein the first image and the second image are captured from distinct viewpoints;

rectifying, by the machine vision system, the first image to form a first rectified image and the second image to form a second rectified image;

retrieving, by the machine vision system, a disparity field that provides mappings from a first plurality of coordinates in the first rectified image to a second plurality of coordinates in the second rectified image;

generating, by the machine vision system, difference data by applying one or more difference functions to a first plurality of image elements in the first rectified image and a corresponding second plurality of image elements in the second rectified image, wherein a correspondence between an element of the first plurality of images and an element of the second plurality of images is based on the disparity field, and wherein the difference function is a function for determining a difference in pixel values, including at least one of;

a difference in grey scale values, a difference in hue values, a difference in gradient values, or a difference in reflectance values; and

determining, by the machine vision system, a presence or an absence of an object on the target surface based on the difference data by applying a filtering algorithm to the difference data, the filtering algorithm comprising setting the plurality of difference values to zero if the plurality of difference values are below a threshold.

9. The method of claim 8, wherein the first image and the second image are captured by a camera and, wherein the camera is configured to image the target surface from the first viewpoint and image the target surface from the second viewpoint approximately simultaneously.

10. The method of claim 8, wherein the first image is captured by a first camera and the second image is captured by a second camera.

11. The method of claim 8, further comprising acquiring the first image and the second image substantially simultaneously.

12. The method of claim 8, wherein the first plurality of image elements is a first plurality of pixel values.

13. The method of claim 8, wherein the target surface is one of a conveyor belt, a tray or a receptacle.

14. The method of claim 8, further comprising applying, by the machine vision system, a shape-finding algorithm to the difference data to determine a presence or an absence of an object on the target surface.

16. The method of claim 14, wherein the shape-finding algorithm comprises blob analysis.