IP Library › Granted Patent US 10,395,341
Granted Patent B2
US 10,395,341 · App. 15/739,801 · Granted Aug 27, 2019

Panoramic image generation method and apparatus for user terminal

Inventors: Guofu Xie (Beijing, CN); Rui Al (Beijing, CN); Li Liu (Beijing, CN); Wenbo Hou (Beijing, CN); Xianpeng Lang (Beijing, CN)
Assignee: BAIDU ONLINE NETWORK TECHNOLOGY (BEIJING) CO., LTD.
G06T3/4038G06K9/6202G06T5/009G06T5/50H04N5/23238
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 10,395,341
App. No.
15/739,801
Granted
Aug 27, 2019
Kind
B2
Abstract

The present disclosure provides a panoramic image generation method and apparatus for a user terminal. The panoramic image generation method for a user terminal includes: obtaining a plurality of images shot by a user terminal, determining an adjacence relationship among the plurality of images, and performing feature matching on adjacent images to obtain matched feature-point-pairs; obtaining an optimized camera parameter according to the matched feature-point-pairs and an initial camera parameter; performing a color adjustment on the adjacent images to obtain color adjusted adjacent images; and mosaicking the color adjusted adjacent images according to the optimized camera parameter to generate a panoramic image.

Claims (73)

1. A panoramic image generation method for a user terminal, comprising:

obtaining a plurality of images shot by a user terminal, determining an adjacence relationship among the plurality of images, and performing feature matching on adjacent images to obtain matched feature-point-pairs;

obtaining an optimized camera parameter according to the matched feature-point-pairs and an initial camera parameter;

performing a color adjustment on the adjacent images to obtain color adjusted adjacent images; and

mosaicking the color adjusted adjacent images according to the optimized camera parameter to generate a panoramic image.

2. The method according to claim 1 , wherein, performing feature matching on adjacent images to obtain matched feature-point-pairs comprises:

dividing each of the adjacent images into a preset number of regions, and extracting feature points with a number less than a preset value from each region;

performing the feature matching according to the feature points extracted from the adjacent images, to obtain the matched feature-point-pairs.

3. The method according to claim 1 , further comprising:

when obtaining the optimized camera parameter is failed, taking a pre-determined ideal camera parameter as the initial camera parameter, and performing an optimization again;

when the optimization is failed again, and each of the adjacent images comprises three layers, removing a bottom layer from each adjacent image and performing the optimization according to two upper layers;

when the optimization according to the two upper layers is failed, performing the optimization utilizing a middle layer of each adjacent image.

4. The method according to claim 1 , wherein, performing a color adjustment on the adjacent images to obtain color adjusted adjacent images comprises:

determining an overlapped area of the adjacent images;

based on two sets of pixel values of pixels in the overlapped area of the adjacent images, determining adjustment parameters minimizing a difference between the two sets of pixel values;

performing the color adjustment on the adjacent images utilizing the adjustment parameters, to obtain the color adjusted adjacent images.

5. The method according to claim 1 , wherein, mosaicking the color adjusted adjacent images according to the optimized camera parameter to generate the panoramic image comprises:

zooming out the color adjusted adjacent images to obtain corresponding zoomed-out images;

determining a mosaic-seam in each zoomed-out image according to the optimized camera parameter;

generating a mask-image corresponding to each of the adjacent image according to the optimized camera parameter and the mosaic-seam, wherein the mask-image comprises a component part and a connection part;

performing a multilayer hybrid fusion on the connection parts of the adjacent images to obtain a fused part, and generating the panoramic image according to the component part and the fused part.

6. The method according to claim 5 , further comprising:

when the component part of the mask-image comprises two parts scattered in two ends of the connection part, dividing the mask-image into two sub images, so that each sub image is processed respectively.

7. A panoramic image generation apparatus for a user terminal, comprising:

one or more processors;

a memory;

one or more software modules stored in the memory and executable by the one or more processors, and comprising:

a matching module, configured to obtain a plurality of images shot by a user terminal, to determine an adjacence relationship among the plurality of images, and to perform feature matching on adjacent images to obtain matched feature-point-pairs;

an optimizing module, configured to obtain an optimized camera parameter according to the matched feature-point-pairs and an initial camera parameter;

an adjusting module, configured to perform a color adjustment on the adjacent images to obtain color adjusted adjacent images; and

a mosaicking module, configured to mosaic the color adjusted adjacent images according to the optimized camera parameter to generate the panoramic image.

8. The apparatus according to claim 7 , wherein, the matching module is configured to perform feature matching on the adjacent images to obtain the matched feature-point-pair by:

dividing each of the adjacent images into a preset number of regions, and extracting feature points with a number less than a preset value from each region;

performing the feature matching according to the feature points extracted from the adjacent images, to obtain the matched feature-point-pairs.

9. The apparatus according to claim 7 , further comprising:

a processing module configured to:

take a pre-determined ideal camera parameter as the initial camera parameter and perform an optimization again when obtaining the optimized camera parameter is failed;

remove a bottom layer from each adjacent image and perform the optimization according to two upper layers when the optimization is failed again and each of the adjacent images comprises three layers; and

perform the optimization utilizing a middle layer of each adjacent image when the optimization based on the two upper layers is failed.

10. The apparatus according to claim 7 , wherein, the adjusting module is configured to:

determine an overlapped area of the adjacent images;

based on two sets of pixel values of pixels in the overlapped area of the adjacent images, determine adjustment parameters minimizing a difference between the two sets of pixel values;

perform the color adjustment on the adjacent images utilizing the adjustment parameters, to obtain the color adjusted adjacent images.

11. The apparatus according to claim 7 , wherein, the mosaicking module is configured to:

zoom out the color adjusted adjacent images to obtain corresponding zoomed-out images;

determine a mosaic-seam in each zoomed-out image according to the optimized camera parameter;

generate a mask-image corresponding to each of the adjacent image according to the optimized camera parameter and the mosaic-seam, wherein the mask-image comprises a component part and a connection part;

perform a multilayer hybrid fusion on the connection parts of the adjacent images to obtain a fused part, and generate the panoramic image according to the component part and the fused part.

12. The apparatus according to claim 11 , wherein, the mosaicking module is configured to:

divide the mask-image into two smaller images when the component part of the mask-image comprises two parts and scattered in two ends of the connection part, so that each smaller image is processed respectively.

13. A non-transitory computer-readable storage medium, having stored therein one or more modules that, when executed, performing a panoramic image generation method for a user terminal, the panoramic image generation method for a user terminal comprising:

obtaining a plurality of images shot by a user terminal, determining an adjacence relationship among the plurality of images, and performing feature matching on adjacent images to obtain matched feature-point-pairs;

obtaining an optimized camera parameter according to the matched feature-point-pairs and an initial camera parameter;

performing a color adjustment on the adjacent images to obtain color adjusted adjacent images; and

mosaicking the color adjusted adjacent images according to the optimized camera parameter to generate a panoramic image.

14. The storage medium according to claim 13 , wherein, performing feature matching on adjacent images to obtain matched feature-point-pairs comprises:

dividing each of the adjacent images into a preset number of regions, and extracting feature points with a number less than a preset value from each region;

performing the feature matching according to the feature points extracted from the adjacent images, to obtain the matched feature-point-pairs.

15. The storage medium according to claim 13 , further comprising:

when obtaining the optimized camera parameter is failed, taking a pre-determined ideal camera parameter as the initial camera parameter, and performing an optimization again;

when the optimization is failed again, and each of the adjacent images comprises three layers, removing a bottom layer from each adjacent image and performing the optimization according to two upper layers;

when the optimization according to the two upper layers is failed, performing the optimization utilizing a middle layer of each adjacent image.

16. The storage medium according to claim 13 , wherein, performing a color adjustment on the adjacent images to obtain color adjusted adjacent images comprises:

determining an overlapped area of the adjacent images;

based on two sets of pixel values of pixels in the overlapped area of the adjacent images, determining adjustment parameters minimizing a difference between the two sets of pixel values;

performing the color adjustment on the adjacent images utilizing the adjustment parameters, to obtain the color adjusted adjacent images.

17. The storage medium according to claim 13 , wherein, mosaicking the color adjusted adjacent images according to the optimized camera parameter to generate the panoramic image comprises:

zooming out the color adjusted adjacent images to obtain corresponding zoomed-out images;

determining a mosaic-seam in each zoomed-out image according to the optimized camera parameter;

generating a mask-image corresponding to each of the adjacent image according to the optimized camera parameter and the mosaic-seam, wherein the mask-image comprises a component part and a connection part;

performing a multilayer hybrid fusion on the connection parts of the adjacent images to obtain a fused part, and generating the panoramic image according to the component part and the fused part.

18. The storage medium according to claim 17 , further comprising:

when the component part of the mask-image comprises two parts scattered in two ends of the connection part, dividing the mask-image into two sub images, so that each sub image is processed respectively.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 30, 2018
From: HOU, WENBO; AI, RUI
To: BAIDU ONLINE NETWORK TECHNOLOGY (BEIJING) CO., LTD.
Reel/Frame 045665/0273 →
Priority Claims (1)
CN 2015 1 0377251 · Jun 30, 2015 · national
Continuity (1)
Related Publication 20180365803A1 · Dec 20, 2018