METHOD FOR RECOGNIZING AND LOCATING OBJECT

• US 20150371102A1
• Filed: 09/24/2014
• Published: 12/24/2015
• Est. Priority Date: 06/18/2014
• Status: Active Grant

• Current Assignee: Delta Electronics Incorporated
• Original Assignee: Delta Electronics Incorporated
• Inventors: Yang, Yuan-Hsiang
• Granted Patent: US 9,396,406 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06T 7/66 of image moments or centre of gravity
• CPC Class Codes: G06T 7/73 using feature-based methods
• CPC Class Codes: G06V 10/243 by compensating for image skew or non-uniform image deformations
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
• CPC Class Codes: G06V 10/40 Extraction of image or video features
• CPC Class Codes: G06V 10/751 Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
• CPC Class Codes: G06V 20/64 Three-dimensional objects

Abstract

A method for recognizing and locating an object includes an offline mode process and an online mode process. In the offline mode process, plural sampled edge points of a template image of the object and respective gradient angles and a gravity position of the plural sampled edge points are obtained, and plural similarity score tables are obtained according to the plural sampled edge points, a predetermined detecting distance range and a predetermined gradient angle difference range. In the online mode process, plural edge points of a live image and respective gradient angles are obtained, plural predictive gravity positions are calculated, and plural similarity scores corresponding to the plural predictive gravity positions are summed up. The predictive gravity position with the local maximum of the similarity scores higher than a threshold value is correlated with the gravity position of the template image so as to recognize and locate the object.

Claims

1. A method for recognizing and locating an object, the method comprising:

performing an offline mode process comprising steps of;

(a) analyzing a template image of the object, so that plural sampled edge points of the template image and respective gradient angles and a gravity position of the plural sampled edge points are obtained; and

(b) obtaining plural similarity score tables according to the plural sampled edge points, a predetermined detecting distance range and a predetermined gradient angle difference range;

performing an online mode process comprising steps of;

(c) analyzing a live image, so that plural edge points of the live image and respective gradient angles of the plural edge points are obtained; and

(d) calculating plural predictive gravity positions corresponding to the plural edge points of the live image, and calculating or summing up plural similarity scores corresponding to the plural predictive gravity positions according to plural similarity scores of the plural similarity score tables, wherein the predictive gravity position with the similarity scores higher than a threshold value is correlated with the gravity position of the template image, thereby recognizing and locating the object.

2. The method according to claim 1, wherein the step (a) further comprises sub-steps of:

(a1) inputting a grayscale image of the template image, and performing a specified edge point capture algorithm, so that plural edge points of the template image are obtained;

(a2) calculating the respective gradient angles of the plural edge points;

(a3) uniformly sampling the plural edge points of the edge point set, so that the plural sampled edge points are obtained; and

(a4) calculating the gravity position of the plural sampled edge points, and a sampled edge point set of the template image is obtained.

3. The method according to claim 1, wherein the step (b) comprises sub-steps of:

(b1) obtaining plural blank similarity score tables according to the plural sampled edge points and the predetermined gradient angle difference range; and

(b2) calculating plural similarity scores according to the plural sampled edge points, the predetermined detecting distance range and the predetermined gradient angle difference range, and filling the plural similarity scores into the plural blank similarity score tables, so that the plural similarity score tables are obtained.

4. The method according to claim 1, wherein the predetermined detecting distance range and the predetermined gradient angle difference range are adjustable parameters.

5. The method according to claim 1, wherein the step (c) comprises sub-steps of:

(c1) inputting a grayscale image of the live image, and performing a specified edge point capture algorithm, so that the plural edge points of the live image are obtained; and

(c2) calculating the respective gradient angles of the plural edge points according to the plural edge points of the live image.

6. The method according to claim 1, wherein the step (d) further comprises a sub-step (d1) of determining plural selected edge points of the live image according to the predetermined gradient angle difference range.

7. The method according to claim 6, wherein in the step (d1), if plural gradient angle differences between the plural edge points of the live image and the plural sampled edge points of the template image are smaller than or equal to the predetermined gradient angle difference range, the plural edge points of the live image are treated as the plural selected edge points of the live image.

8. The method according to claim 6, wherein after the sub-step (d1), the step (d) further comprises a sub-step (d2) of obtaining the plural predictive gravity positions according to comparisons between the plural selected edge points of the live image and the plural sampled edge points of the template image.

9. The method according to claim 8, wherein after the sub-step (d2), the step (d) further comprises a sub-step (d3) of selecting the similarity score tables corresponding to the plural predictive gravity positions according to the plural gradient angle differences between the plural selected edge points of the live image and the plural sampled edge points of the template image.

10. The method according to claim 9, wherein after the sub-step (d3), the step (d) further comprises a sub-step (d4) of calculating the plural similarity scores corresponding to the plural predictive gravity positions according to the corresponding similarity score tables.

11. The method according to claim 10, wherein in the sub-step (d4), when the plural similarity scores corresponding to the plural predictive gravity positions are calculated and if two of the plural predictive gravity positions have plural identical similarity scores corresponding to plural similarity score tables, a method of calculating the plural similarity scores is determined according to whether the two predictive gravity positions are obtained by using the same sampled edge point as a reference point.

12. The method according to claim 11, wherein in the sub-step (d4), if the two predictive gravity positions are obtained by using the same sampled edge point as the reference point, the similarity scores of the selected edge points are replaced by merit and the higher similarity scores of the plural similarity score tables corresponding to the predictive gravity positions are determined as the similarity score of respective predictive gravity positions.

13. The method according to claim 11, wherein in the sub-step (d4), if the two predictive gravity positions are obtained by using the different sampled edge points as the reference point, the similarity scores of the plural similarity score tables corresponding to the predictive gravity positions are summed up and determined as the similarity scores of the predictive gravity positions.

14. The method according to claim 1, wherein the threshold value is set and adjustable by a user, and the predictive gravity position with a local maximum of the similarity scores higher than a threshold value is correlated with the gravity position of the template image.