Method and mobile terminal for previewing and retrieving video

• US 8,737,808 B2
• Filed: 10/18/2010
• Issued: 05/27/2014
• Est. Priority Date: 05/14/2010
• Status: Active Grant

• Current Assignee: ZTE Corporation
• Original Assignee: ZTE Corporation
• Inventors: Gao, Minggang
• Primary Examiner(s): HARVEY, DAVID E
• Application Number:
  US13/575,956
• Publication Number:
  US 20120308205A1
• Time in Patent Office: 1,317 Days
• Field of Search: 386/343, 386/344, 386/241, 386/248, 715/721, 715/723
• US Class Current: 386/241
• CPC Class Codes: G06F 16/7328 Query by example, e.g. a complete video frame or video sequence graphical querying G06F16/7335
• CPC Class Codes: G06F 16/785 using colour or luminescence
• CPC Class Codes: G06V 20/49 Segmenting video sequences, i.e. computational techniques such as parsing or cutting the sequence, low-level clustering or determining units such as shots or scenes

Abstract

A method for previewing and retrieving a video is disclosed. The method includes: (101) setting a variable-length buffer sequence; (102) acquiring and processing a video file to obtain images; (103) detecting scene shots in the obtained images and key frames of the scene shots, and storing the detected scene shots and key frames into the variable-length buffer sequence; and (104) acquiring the scene shots and key frames from the variable-length buffer sequence, matching and comparing the scene shots and key frames with a specified target image to perform image retrieval. A mobile terminal for previewing and retrieving a video is further disclosed. With the technical solution, content preview and fast retrieval of a video file are achieved in a mobile terminal with relatively limited processing capacity and memory resources.

Claims

1. A method for previewing and retrieving a video, comprising:

setting a variable-length buffer sequence; and

further comprising;

acquiring and processing a video file to obtain images;

detecting scene shots in the obtained images and key frames of the scene shots, and storing the detected scene shots and key frames in the variable-length buffer sequence; and

acquiring the scene shots and key frames from the variable-length buffer sequence, and performing video retrieval by matching and comparing the scene shots and key frames with a specified target image,

wherein the performing video retrieval by matching and comparing the scene shots and key frames with the target image is;

calculating degrees of match between the scene shots and the target image, and degrees of match between the key frames and the target image;

determining whether the degrees of match between the scene shots and the target image reach a threshold of degree of match;

when none of the degrees of match between the scene shots and the target image reaches the threshold of degree of match, determining whether the degrees of match between the key frames and the target image reach the threshold of degree of match;

when none of the degrees of match between the key frames and the target image reaches the threshold of degree of match, determining that the matching fails, and ending the video retrieval;

when a degree of match between a key frame and the target image reaches the threshold of degree of match, determining that the matching succeeds, ending the video retrieval, and executing a notification mechanism of successful retrieval.

2. The method according to claim 1, wherein the setting the variable-length buffer sequence is:

applying from a Central Processing Unit (CPU) of a mobile terminal for a dynamic memory area in a memory of the mobile terminal, and taking the dynamic memory area as the variable-length buffer sequence.

3. The method according to claim 1, wherein the detecting the scene shots in the obtained images is:

setting a sliding window, taking a first frame of the images as a first scene shot, adjusting a frame, which is to be detected and falls in the sliding window, to be at the center of the sliding window;

calculating chromaticity histogram difference between the first frame of the images and the frame to be detected, and confirming the frame to be detected as a scene shot when the chromaticity histogram difference meets a certain condition.

4. The method according to claim 3, wherein the certain condition is:

the frame has a greatest chromaticity histogram difference, and the greatest chromaticity histogram difference is two or three times greater than a second greatest chromaticity histogram difference.

5. The method according to claim 2, wherein detecting the key frames of the scene shots is:

sorting frames between the first frame and a detected scene shot according to magnitude of chromaticity histogram differences, and taking three frames with the greatest chromaticity histogram differences as the key frames of the scene shot.

6. The method according to claim 1, wherein the determining whether the degrees of match between the scene shots and the target image reach the threshold of degree of match is:

setting the threshold of degree of match, randomly selecting a scene shot from all the scene shots, determining whether the degree of match between the scene shot and the target image reaches the threshold of degree of match, if not, selecting another scene shot till all the scene shots are determined or till a scene shot with a degree of match which reaches the threshold of degree of match is found.

7. The method according to claim 1, wherein calculating the degrees of match between the scene shots and the target image, and the degrees of match between the key frames and the target image is:

dividing the image of the scene shots or key frames into zones, calculating an average Red-Green-Blue (RGB) value of each zone, calculating a difference between the average RGB value of each zone and an RGB value of a corresponding zone in the target image, summing over all the differences and performing normalization on the sum to obtain degree of image difference of each zone, and summing over the degrees of image difference of all the zones and calculating an average value of degree of image difference.

8. The method according to claim 1, wherein the determining whether the degrees of match between the key frames and the target image reach the threshold of degree of match is:

when none of the degrees of match between the scene shots and the target image reaches the threshold, prioritizing the scene shots in descending order of the degrees of match, determining whether the degree of match between any key frame of the scene shot with the top priority and the target image reaches the threshold of degree of match;

if not, determining whether the degrees of match between the other key frames of the scene shot and the target image reach the threshold of degree of match;

if none of the degrees of match between the three key frames of the scene shot with the top priority and the target image reaches the threshold of degree of match, determining whether the degrees of match between the key frames of the scene shot with the second highest priority and the target image reach the threshold of degree of match.

9. The method according to claim 1, wherein the executing the notification mechanism of successful retrieval is:

presetting a shared variable, setting the shared variable to be true when the matching succeeds, determining whether the shared variable is set to be true before each detection of a new scene shot or key frame, ending the detection when the shared variable is set to be true, otherwise continuing the detection.

10. The method according to claim 3, wherein detecting the key frames of the scene shots is:

sorting frames between the first frame and a detected scene shot according to magnitude of chromaticity histogram differences, and taking three frames with the greatest chromaticity histogram differences as the key frames of the scene shot.

11. The method according to claim 4, wherein detecting the key frames of the scene shots is:

sorting frames between the first frame and a detected scene shot according to magnitude of chromaticity histogram differences, and taking three frames with the greatest chromaticity histogram differences as the key frames of the scene shot.

12. A mobile terminal for previewing and retrieving a video, the mobile terminal comprising one or more processors for implementing:

an acquisition module is configured to acquire and process a video file to obtain images and send the obtained images to the detection module;

a detection module is configured to detect scene shots in the obtained images and key frames of the scene shots, and store the detected scene shots and key frames into the variable-length buffer sequence;

a variable-length buffer sequence is configured to store the scene shots and key frames sent by the detection module; and

a video retrieval module is configured to acquire the scene shots and key frames from the variable-length buffer sequence, and match and compare the scene shots and key frames with a specified target image to perform video retrieval,

wherein the performing video retrieval by matching and comparing the scene shots and key frames with the target image is;

calculating degrees of match between the scene shots and the target image, and degrees of match between the key frames and the target image;

determining whether the degrees of match between the scene shots and the target image reach a threshold of degree of match;

when none of the degrees of match between the scene shots and the target image reaches the threshold of degree of match, determining whether the degrees of match between the key frames and the target image reach the threshold of degree of match;

when none of the degrees of match between the key frames and the target image reaches the threshold of degree of match, determining that the matching fails, and ending the video retrieval;

when a degree of match between a key frame and the target image reaches the threshold of degree of match, determining that the matching succeeds, ending the video retrieval, and executing a notification mechanism of successful retrieval.

13. The mobile terminal according to claim 12, wherein the one or more processors are further configured to implement a setting module, and the setting module is configured to set the variable-length buffer sequence and a threshold of degree of match.