# US 10,122,916 B2 · Object monitoring method and device

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 10,122,916 B2
- Document ID: US10122916B2
- Status: Granted
- Kind: B2
- Application number: 15/378,059
- Filed: 2016-12-14
- Published: 2017-10-26
- Granted: 2018-11-06
- Earliest claimed: 2016-04-22
- Assignees: Beijing Xiaomi Mobile Software Co., Ltd.
- Applicants: Beijing Xiaomi Mobile Software Co., Ltd.
- Inventors: Weiguang Jia; Na Ma; Enxing Hou
- Web page: https://patentgrep.com/patent/US10122916B2

## CPC

- H04N 23/64: {Computer-aided capture of images, e.g. transfer from script file into camera, check of taken image quality, advice or proposal for image composition or decision on when to take image}
- G08B 13/19663: {Surveillance related processing done local to the camera}
- G08B 13/19682: {Graphic User Interface [GUI] presenting system data to the user, e.g. information on a screen helping a user interacting with an alarm system}
- G08B 13/19684: {Portable terminal, e.g. mobile phone, used for viewing video remotely}
- G08B 13/19689: {Remote control of cameras, e.g. remote orientation or image zooming control for a PTZ camera}
- G08B 13/19697: {Arrangements wherein non-video detectors generate an alarm themselves}
- G08B 21/0208: {Combination with audio or video communication, e.g. combination with "baby phone" function}
- G08B 25/08: using communication transmission lines {(G08B13/19658, G08B21/0286, G08B25/016 take precedence)}
- G10L 15/02: Feature extraction for speech recognition; Selection of recognition unit
- G10L 15/063: {Training}
- G10L 25/51: for comparison or discrimination
- H04N 7/188: {Capturing isolated or intermittent images triggered by the occurrence of a predetermined event, e.g. an object reaching a predetermined position (signal generation from motion picture films H04N5/253)}
- H04N 23/63: by using electronic viewfinders
- H04N 23/661: Transmitting camera control signals through networks, e.g. control via the Internet
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION
- G08B: SIGNALLING SYSTEMS, e.g. PERSONAL CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING

## Abstract

An object monitoring method and device are provided. The object monitoring method includes: a sound production reason for the target sound is determined when a target sound exists in acquired sounds; and sound production information is pushed to a mobile terminal, the sound production information comprising the sound production reason.

## Claims

### Claim 1

An object monitoring method, comprising:
determining a sound production reason for a target sound when the target sound exists in acquired sounds;
pushing sound production information to a mobile terminal, the sound production information comprising the sound production reason;
receiving a video acquisition instruction sent by the mobile terminal;
pushing video information to the mobile terminal according to the video acquisition instruction so as to enable the mobile terminal to display the video information;
broadcasting a message to enable the mobile terminal to display prompt information indicating that a current monitoring equipment does not access a network responsive to detecting that the current monitoring equipment does not access the network;
receiving a name and a password of a router broadcast by the mobile terminal; and
sending a connection request to the router for connection to the router.

### Claim 2 (depends on 1)

The object monitoring method according to claim 1, further comprising: before pushing the video information to the mobile terminal according to the video acquisition instruction,
acquiring a sound production position of the target sound; and
controlling a camera to turn to a direction corresponding to the sound production position and acquire the video information.

### Claim 3 (depends on 1)

The object monitoring method according to claim 1, wherein determining the sound production reason for the target sound comprises:
performing feature extraction on the target sound to obtain feature information of the target sound; and
matching the feature information of the target sound with a plurality of sound templates which are pre-stored, and obtaining the sound production reason for the target sound according to a matching result, the plurality of sound templates being configured to identify target sounds produced for different reasons.

### Claim 4 (depends on 3)

The object monitoring method according to claim 3, wherein performing feature extraction on the target sound to obtain the feature information of the target sound comprises:
compressing the target sound to obtain compressed sound information; and
performing feature extraction on the compressed sound information to obtain the feature information of the target sound.

### Claim 5 (depends on 1)

The object monitoring method according to claim 1, wherein determining the sound production reason for the target sound comprises:
sending the target sound to a server; and
receiving the sound production reason for the target sound returned by the server.

### Claim 6 (depends on 3)

The object monitoring method according to claim 3, further comprising:
training a current sound template, wherein each of the plurality of sound templates is taken as the current sound template respectively.

### Claim 7 (depends on 6)

The object monitoring method according to claim 6, wherein training the current sound template comprises:
acquiring a preset number of target sound feature samples;
inputting the preset number of target sound feature samples into the current sound template to obtain a distortion distance of each of the preset number of target sound feature samples relative to the current sound template, the distortion distance representing a difference between each of the preset number of target sound feature samples and a target sound output through the current sound template;
stopping training of the current sound template responsive to the distortion distance being less than a preset distance;
continuing training of the current sound template until the distortion distance is less than the preset distance responsive to the distortion distance being greater than or equal to the preset distance.

### Claim 8 (depends on 1)

The object monitoring method according to claim 1, wherein the sound production information further comprises the target sound.

### Claim 9

An object monitoring device, comprising:
a processor; and
a memory for storing instructions executable by the processor,
wherein the processor is configured to:
determine a sound production reason for a target sound when the target sound exists in acquired sounds;
push sound production information to a mobile terminal, the sound production information comprising the sound production reason;
receive a video acquisition instruction sent by the mobile terminal;
push video information to the mobile terminal according to the video acquisition instruction so as to enable the mobile terminal to display the video information;
broadcast a message to enable the mobile terminal to display prompt information indicating that a current monitoring equipment does not access a network responsive to detecting that the current monitoring equipment does not access the network;
receive a name and a password of a router broadcast by the mobile terminal; and
send a connection request to the router for connection to the router.

### Claim 10 (depends on 9)

The object monitoring device according to claim 9, wherein the processor is further configured to, before pushing the video information to the mobile terminal according to the video acquisition instruction,
acquire a sound production position of the target sound; and
control a camera to turn to a direction corresponding to the sound production position and acquire the video information.

### Claim 11 (depends on 9)

The object monitoring device according to claim 9, wherein in order to determine the sound production reason for the target sound, the processor is configured to:
perform feature extraction on the target sound to obtain feature information of the target sound; and
match the feature information of the target sound with a plurality of sound templates which are pre-stored, and obtain the sound production reason for the target sound according to a matching result, the plurality of sound templates being configured to identify target sounds produced for different reasons.

### Claim 12 (depends on 11)

The object monitoring device according to claim 11, wherein in order to perform feature extraction on the target sound to obtain the feature information of the target sound, the processor is configured to:
compress the target sound to obtain compressed sound information; and
perform feature extraction on the compressed sound information to obtain the feature information of the target sound.

### Claim 13 (depends on 9)

The object monitoring device according to claim 9, wherein in order to determine the sound production reason for the target sound, the processor is configured to:
send the target sound to a server; and
receive the sound production reason, returned by the server, for the target sound.

### Claim 14 (depends on 11)

The object monitoring device according to claim 11, wherein the processor is further configured to:
train a current sound template, wherein each of the plurality of sound templates is taken as the current sound template respectively.

### Claim 15 (depends on 14)

The object monitoring device according to claim 14, wherein in order to train the current sound template, the processor is configured to:
acquire a preset number of target sound feature samples;
input the preset number of target sound feature samples into the current sound template to obtain a distortion distance of the preset number of target sound feature samples relative to the current sound template, the distortion distance representing a difference between the preset number of target sound feature samples and a target sound output through the current sound template;
stop training of the current sound template responsive to the distortion distance being less than a preset distance;
continue training of the current sound template until the distortion distance is smaller than the preset distance responsive to the distortion distance being greater than or equal to the preset distance.

### Claim 16

A non-transitory computer-readable storage medium having stored therein instructions that, when executed by a processor, causes the processor to perform an object monitoring method, the method comprising:
determining a sound production reason for a target sound when the target sound exists in acquired sounds;
pushing sound production information to a mobile terminal, the sound production information comprising the sound production reason;
receiving a video acquisition instruction sent by the mobile terminal;
pushing video information to the mobile terminal according to the video acquisition instruction so as to enable the mobile terminal to display the video information;
broadcasting a message to enable the mobile terminal to display prompt information indicating that a current monitoring equipment does not access a network responsive to detecting that the current monitoring equipment does not access the network;
receiving a name and a password of a router broadcast by the mobile terminal; and
sending a connection request to the router for connection to the router.