Alarm method and device

• US 10,147,288 B2
• Filed: 03/17/2016
• Issued: 12/04/2018
• Est. Priority Date: 10/28/2015
• Status: Active Grant

• Current Assignee: Xiaomi Inc. (Xiaomi Corp.)
• Original Assignee: Xiaomi Inc. (Xiaomi Corp.)
• Inventors: Zhang, Tao, Chen, Zhijun, Wang, Pingze
• Primary Examiner(s): Williams, Jeffery A
• Application Number:
  US15/072,827
• Publication Number:
  US 20170124821A1
• Time in Patent Office: 992 Days
• Field of Search: 348155
• CPC Class Codes: G06F 18/214 Generating training patterns; Bootstrap methods, e.g. bagging or boosting
• CPC Class Codes: G06F 18/217 Validation; Performance evaluation; Active pattern learning techniques
• CPC Class Codes: G08B 13/19606 Discriminating between target movement or movement in an area of interest and other non-signicative movements, e.g. target movements induced by camera shake or movements of pets, falling leaves, rotating fan
• CPC Class Codes: G08B 13/19656 Network used to communicate with a camera, e.g. WAN, LAN, Internet
• CPC Class Codes: G08B 13/1966 Wireless systems, other than telephone systems, used to communicate with a camera

Abstract

An alarm method is disclosed. The method may comprise acquiring a monitoring video, detecting whether the monitoring video contains a moving target, determining whether the moving target is a preset object if it is detected that the monitoring video contains the moving target, and forbidding sending corresponding alarm information to a terminal if the moving target is determined to be the preset object.

Claims

1. A processor-implemented alarm method, comprising:

acquiring a monitoring video from camera equipment in communication with a processor;

detecting whether the monitoring video contains a moving target;

if it is detected that the monitoring video contains the moving target, determining whether the moving target is a preset object; and

if the moving target is determined to be the preset object, forbidding sending corresponding alarm information to a terminal,

wherein detecting whether the monitoring video contains the moving target comprises;

acquiring a pixel value of each pixel in a video image of the monitoring video;

determining, based on the pixel value of each pixel in the video image and a preset background model, whether the video image contains foreground pixels, the preset background model being configured to represent a distribution characteristic of a pixel value of each background pixel in the video image on a time domain; and

if it is determined that the video image contains the foreground pixels, determining that the monitoring video contains the moving target; and

wherein determining whether the moving target is a preset object comprises;

in the video image of the monitoring video, cropping an area where the moving target is located to obtain a target image;

processing a size of the target image into a preset size;

extracting a characteristic vector from the target image according to a preset classification model;

calculating probabilities of the target image belonging to multiple types included in the preset classification model, according to the characteristic vector and the preset classification model;

determining a type with a maximum probability as a type of the moving target; and

when the determined type is a preset type, determining that the moving target is the preset object.

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

updating the preset background model based on the pixel value of each pixel in the video image.

3. The method according to claim 1, wherein the preset object comprises a pet.

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

pre-establishing the preset classification model by acquiring sample picture sets corresponding to multiple types respectively;

wherein the preset classification model is configured to distinguish the multiple types and determine a type of an image.

11. The method according to claim 1, wherein determining, based on the pixel value of each pixel in the video image and a preset background model, whether the video image contains foreground pixels comprises:

matching the pixel value of each pixel with the preset background model;

determining that the video image does not contain the foreground pixel when the pixel value of each pixel is matched with the preset background model;

otherwise, determining that the video image contains the foreground pixel;

wherein the foreground pixel is a pixel with a pixel value unsuccessfully matched with the preset background model.

4. An alarm device, comprising:

a processor in communication with camera equipment; and

a memory configured to store instructions executable by the processor,

wherein the processor is configured to;

acquire a monitoring video from the camera equipment;

detect whether the monitoring video contains a moving target;

if it is detected that the monitoring video contains the moving target, determine whether the moving target is a preset object; and

if the moving target is determined to be the preset object, forbid sending corresponding alarm information to a terminal,

wherein, in detecting whether the monitoring video contains the moving target, the processor is further configured to;

acquire a pixel value of each pixel in a video image of the monitoring video;

determine, based on the pixel value of each pixel in the video image and a preset background model, whether the video image contains foreground pixels, the preset background model being configured to represent a distribution characteristic of a pixel value of each background pixel in the video image on a time domain; and

if it is determined that the video image contains the foreground pixels, determine that the monitoring video contains the moving target; and

wherein, in determining whether the moving target is a preset object, the processor is further configured to;

in the video image of the monitoring video, crop an area where the moving target is located to obtain a target image;

process a size of the target image into a preset size;

extract a characteristic vector from the target image according to a preset classification model;

calculate probabilities of the target image belonging to multiple types included in the preset classification model, according to the characteristic vector and the preset classification model;

determine a type with a maximum probability as a type of the moving target; and

when the determined type is a preset type, determine that the moving target is the preset object.

5. The device of claim 4, wherein the processor is further configured to:

update the preset background model based on the pixel value of each pixel in the video image.

6. The device of claim 4, wherein the preset object comprises a pet.

7. A non-transitory computer-readable storage medium storing instructions that, when executed by a processor of a device, cause the device to perform an alarm method, the method comprising:

acquiring a monitoring video from camera equipment in communication with the processor;

detecting whether the monitoring video contains a moving target;

if it is detected that the monitoring video contains the moving target, determining whether the moving target is a preset object; and

if it is determined that the moving target is the preset object, forbidding sending corresponding alarm information to a terminal,

wherein detecting whether the monitoring video contains the moving target comprises;

acquiring a pixel value of each pixel in a video image of the monitoring video;

determining, based on the pixel value of each pixel in the video image and a preset background model, whether the video image contains foreground pixels, the preset background model being configured to represent a distribution characteristic of a pixel value of each background pixel in the video image on a time domain; and

if it is determined that the video image contains the foreground pixels, determining that the monitoring video contains the moving target; and

wherein determining whether the moving target is a preset object comprises;

in a video image of the monitoring video, cropping an area where the moving target is located to obtain a target image;

processing a size of the target image into a preset size;

extracting a characteristic vector from the target image according to a preset classification model;

calculating probabilities of the target image belonging to multiple types included in the preset classification model, according to the characteristic vector and the preset classification model;

determining a type with a maximum probability as a type of the moving target; and

when the determined type is a preset type, determining that the moving target is the preset object.

8. The non-transitory computer-readable storage medium of claim 7, the method further comprising:

updating the preset background model based on the pixel value of each pixel in the video image.

9. The non-transitory computer-readable storage medium of claim 7, wherein the preset object comprises a pet.