IP Library › Granted Patent US 11,893,718
Granted Patent B2
US 11,893,718 · App. 17/623,662 · Granted Feb 6, 2024

Image recognition method and image recognition device

Inventors: Seiji Yamagami (Kyoto, JP); Munekuni Oshima (Ise, JP)
Assignee: MURATA MACHINERY, LTD.
G06T5/50G06T7/174G06V10/7715G06V20/59G06T2207/20224G06V2201/07
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,893,718
App. No.
17/623,662
Granted
Feb 6, 2024
Kind
B2
Abstract

A method of recognizing a specific image includes detecting image features from the specific image and storing in a memory each of detected relevant image features, detecting relevant image features from an input image and selecting from the memory a portion corresponding to each of the detected relevant image features, generating a restoration image using portions selected in the selecting, and determining whether the generated restoration image matches with the input image by a matching process and recognizing that, when it is determined that the restoration image matches with the input image, the input image is the specific image.

Claims (37)

1. An image recognition method of recognizing a specific image, the method comprising:

a first step of detecting a plurality of relevant image features from the specific image and storing in a memory each of the detected relevant image features as a portion;

a second step of detecting a plurality of image features from an input image and selecting from the memory the portion corresponding to each of the detected relevant image features;

a third step of generating a restoration image by using a plurality of portions selected in the second step; and

a fourth step of determining whether the restoration image generated in the third step matches with the input image by a matching process and recognizing that, when it is determined that the restoration image matches with the input image, the input image is the specific image.

2. The image recognition method according to claim 1 , wherein

in the first step, a portion label as a label is given to each of the portions, and the portion label is stored in the memory together with a corresponding one of the portions; and

in the second step, a selection label as a label is given to each of the relevant image features, and the portion of the portion label that matches with the selection label is selected.

3. The image recognition method according to claim 1 , wherein

the second step includes an image cutting process of cutting out an input image from a captured image as a distance image; and

in the image cutting process, a point cloud of a depth distance within a predetermined range in the captured image is assumed as a relevant object, and an image of the relevant object in the captured image is cut out as the input image.

4. The image recognition method according to claim 1 , wherein

in the second step, a feature intensity of each of the image features is detected; and

in the third step, the restoration image is generated by further using the plurality of feature intensities detected in the second step.

5. The image recognition method according to claim 1 , wherein the specific image is an image of at least a portion of a vehicle body in a traveling vehicle traveling along a predetermined traveling path.

6. An image recognition device to recognize a specific image, the image recognition device comprising:

a memory to store therein each of a plurality of image features detected from the specific image as a portion in advance;

a feature detector to detect a plurality of relevant image features from an input image;

a restorer to select from the memory the portion corresponding to each of the relevant image features detected by the feature detector and to generate a restoration image by using a plurality of selected portions; and

a determiner to determine whether the restoration image generated in the restorer matches with the input image by a matching process and to recognize that, when it is determined that the restoration image matches with the input image, the input image is the specific image.

7. The image recognition device according to claim 6 , wherein

the memory stores therein a label given to each of a plurality of portions as a portion label together with the portion in advance;

the feature detector gives a selection label as a label to each of the image features; and

the restorer selects the portion of the portion label that matches with the selection label.

8. The image recognition device according to claim 6 further comprising:

an image cutter to cut out the input image from a captured image as a distance image; wherein

the image cutter assumes as a relevant object a point cloud of a depth distance within a predetermined range in the captured image and cuts out an image of the relevant object in the captured image as the input image.

9. The image recognition device according to claim 6 , wherein

the feature detector detects a feature intensity of each of the image features; and

the restorer generates the restoration image by further using the plurality of feature intensities detected by the feature detector.

10. The image recognition device according to claim 6 , wherein the specific image is an image of at least a portion of a vehicle body in a traveling vehicle traveling along a predetermined traveling path.

11. The image recognition method according to claim 1 , wherein, in the first step, the detected relevant image features are detected from the specific image by passing the specific image through an image filter.

12. The image recognition method according to claim 1 , wherein, in the second step, the plurality of image features are detected from the input image by passing the input image through an image filter.

13. The image recognition method according to claim 1 , wherein the matching process of the fourth step includes calculating a similarity of the restoration image to the input image and determining that, when the similarity is greater than or equal to a threshold value, the restoration image matches with the input image.

14. The image recognition device according to claim 6 , further comprising an image filter to detect plurality of image features from the specific image.

15. The image recognition device according to claim 6 , wherein the feature detector includes an image filter to detect the plurality of image features from the input image.

16. The image recognition device according to claim 6 , wherein the matching process of the determiner includes calculating a similarity of the restoration image to the input image and determining that, when the similarity is greater than or equal to a threshold value, the restoration image matches with the input image.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 29, 2021
From: YAMAGAMI, SEIJI; OSHIMA, MUNEKUNI
To: MURATA MACHINERY, LTD.
Reel/Frame 058498/0842 →
Priority Claims (1)
JP 2019-130264 · Jul 12, 2019 · national
Continuity (1)
Related Publication 20220253994A1 · Aug 11, 2022