FOCUSING POINT DETERMINING METHOD AND APPARATUS

• US 20170374268A1
• Filed: 02/27/2017
• Published: 12/28/2017
• Est. Priority Date: 06/28/2016
• Status: Active Grant

• Current Assignee: Beijing Kuangshi Technology Co., Ltd., Pinhole (Beijing) Technology Co., Ltd.
• Original Assignee: Beijing Kuangshi Technology Co., Ltd., Pinhole (Beijing) Technology Co., Ltd.
• Inventors: ZHOU, Shuchang, YAO, Cong, ZHOU, Xinyu, HE, Weiran, FENG, Dieqiao, YIN, Qi
• Granted Patent: US 10,291,838 B2
• Time in Patent Office: Days
• CPC Class Codes: G06V 10/25 Determination of region of interest [ROI] or a volume of interest [VOI]
• CPC Class Codes: H04N 23/675 comprising setting of focusing regions
• CPC Class Codes: H04N 23/80 Camera processing pipelines; Components thereof

Abstract

There are provided a focusing point determining method and apparatus. The focusing point determining method comprises: obtaining a view-finding image within a view-finding coverage; identifying a significance area in the view-finding image; and extracting at least one focusing point from the identified significance area. By identifying the significance area in the view-finding image and extracting at least one focusing point from the identified significance area, the focusing point determining method and apparatus can ensure accuracy of a selected focusing point to a certain extent, so as to ensure accuracy of focusing.

Claims

1. A focusing point determining method, comprising:

obtaining a view-finding image within a view-finding coverage;

identifying a significance area in the view-finding image; and

extracting at least one focusing point from the identified significance area.

2. The focusing point determining method according to claim 1, wherein the significance area in the view-finding image is identified by utilizing neural networks.

3. The focusing point determining method according to claim 1, wherein said identifying the significance area in the view-finding image comprises:

generating a significance identifying picture based on the view-finding image, wherein in the significance identifying picture, a gray scale value of each pixel represents significance of the pixel;

binarizing the significance identifying picture to obtain a binarized significance picture; and

taking at least one connected area in the binarized significance picture as the significance area.

4. The focusing point determining method according to claim 3, wherein said generating a significance identifying picture based on the view-finding image comprises:

calculating an image domain gradient of the view-finding image by using neural networks; and

generating the significance identifying picture according to the image domain gradient.

5. The focusing point determining method according to claim 3, wherein said extracting at least one focusing point from the identified significance area comprises:

extracting one focusing point for each connected area in the binarized significance picture.

6. The focusing point determining method according to claim 5, wherein said extracting one focusing point for each connected area in the binarized significance picture comprises:

calculating an average value of coordinates of respective points in the connected area, and taking the average value as a coordinate of the focusing point corresponding to the connected area.

7. The focusing point determining method according to claim 1, wherein said extracting at least one focusing point from the identified significance area comprises:

extracting at least one predicted focusing point from the identified significance area;

for each predicted focusing point, searching a historical focusing point corresponding to the predicted focusing point in a neighborhood of a predetermined radius which takes the predicted focusing point as a center, the historical focusing point being the focusing point based on which photographing is performed last time; and

for each predicted focusing point, in the case of finding a historical focusing point corresponding to the predicted focusing point, performing smoothing process on the predicted focusing point by using the historical focusing point corresponding to the predicted focusing point, so as to obtain a focusing point corresponding to the predicted focusing point.

8. The focusing point determining method according to claim 7, wherein said extracting at least one focusing point from the identified significance area comprises:

extracting one predicted focusing point for each connected area in the binarized significance picture.

9. The focusing point determining method according to claim 8, wherein said extracting one focusing point for each connected area in the binarized significance picture comprises:

calculating an average value of coordinates of respective points in the connected area, and taking the average value as a coordinate of the predicted focusing point corresponding to the connected area.

10. The focusing point determining method according to claim 1, further comprising:

performing focusing operation on the view-finding coverage by using the extracted focusing point and photographing an image.

11. A focusing point determining apparatus, comprising:

an image obtaining means, configured to obtain a view-finding image within a view-finding coverage;

a significance area identifying means, configured to identify a significance area in the view-finding image; and

a focusing point extracting means, configured to extract at least one focusing point in the identified significance area.

12. The focusing point determining apparatus according to claim 11, wherein the significance area identifying means identifies the significance area in the view-finding image by utilizing neural networks.

13. The focusing point determining apparatus according to claim 11, wherein the significance area identifying means comprises:

a significance picture generating means, configured to generate a significance identifying picture based on the view-finding image, wherein in the significance identifying picture, a gray scale value of each pixel represents significance of the pixel;

a binarization means, configured to binarize the significance identifying picture to obtain a binarized significance picture; and

a significance area extracting means configured to take at least one connected area in the binarized significance picture as a significance area.

14. The focusing point determining apparatus according to claim 13, wherein significance picture generating means calculates an image domain gradient of the view-finding image by using neural networks, and generates the significance identifying picture according to the image domain gradient.

15. The focusing point determining apparatus according to claim 13, wherein for each connected area in the binarized significance picture, the focusing point extracting means extracts one focusing point.

16. The focusing point determining apparatus according to claim 15, wherein for each connected area in the binarized significance picture, the focusing point extracting means calculates an average value of coordinates of respective points in the connected area, and takes the average value as a coordinate of the focusing point corresponding to the connected area.

17. The focusing point determining apparatus according to claim 11, wherein the focusing point extracting means comprises:

a predicted focusing point extracting means, configured to extract at least one predicted focusing point from the identified significance area;

a historical focusing point searching means, configured to for each predicted focusing point, search a historical focusing point corresponding to the predicted focusing point in a neighborhood of a predetermined radius which takes the predicted focusing point as a center, the historical focusing point being the focusing point based on which photographing is performed last time; and

a smoothing process means, configured to for each predicted focusing point, in the case of finding a historical focusing point corresponding to the predicted focusing point, perform smoothing process on the predicted focusing point by using the historical focusing point corresponding to the predicted focusing point, so as to obtain a focusing point corresponding to the predicted focusing point.

18. The focusing point determining apparatus according to claim 17, wherein for each connected area in the binarized significance picture, the predicted focusing point extracting means extracts one predicted focusing point.

19. The focusing point determining apparatus according to claim 18, wherein for each connected area in the binarized significance picture, the predicted focusing point extracting means calculates an average value of coordinates of respective points in the connected area, and takes the average value as a coordinate of the predicted focusing point corresponding to the connected area.

20. The focusing point determining apparatus according to claim 11, further comprising:

performing focusing operation on the view-finding coverage by using the extracted focusing point and photographing an image.