KNOCK DETECTION USING A DOORBELL CAMERA
Presented herein are systems and methods for a camera that can detect a knock. A system can include a doorbell housing. The doorbell housing can include a camera, microphone, and processor. The camera can capture video data of a detection zone in front of a door. The microphone can receive audio. The processor can be a processor coupled to the camera and microphone. The processor can detect a knock in the audio and generate a notification to transmit to a homeowner. The processor can activate the microphone upon detecting a person within the detection zone using the camera. The microphone can capture audio within the detection zone.
1 . A system, comprising:
a doorbell housing comprising:
a camera configured to capture video data of a detection zone in front of a door;
a microphone configured to receive audio data;
a processor coupled to the camera and microphone, the processor configured to:
detect a knock in the audio data; and
generate a notification to transmit to a homeowner.
2 . The system of claim 1 , wherein the processor is configured to activate the microphone to receive the audio data upon detecting a person within the detection zone using the camera.
3 . The system of claim 1 , wherein the microphone is configured to capture the audio data within the detection zone.
4 . The system of claim 1 , wherein processor is further configured to analyze the video data to identify movement in the detection zone.
5 . The system of claim 1 , wherein the processor executes instructions corresponding to a model trained on at least one of audio or visual data.
6 . The system of claim 1 , wherein the processor is configured to distinguish between human and non-human presence in the detection zone based on at least one of the audio data or video data.
7 . The system of claim 1 , wherein the notification facilitates interaction of the homeowner to the detection zone.
8 . A method, comprising:
capturing video data from a detection zone in front of a door;
capturing audio data from in front of the door;
detecting a knock based on the audio data; and
generating a notification based on the detected knock; and
transmitting the notification to a homeowner indicative of the knock.
9 . The method of claim 8 , further comprising activating a microphone to capture the audio data upon detection of a person based on the video data.
10 . The method of claim 8 , wherein the audio data is captured by a microphone of a doorbell positioned exterior to a front door.
11 . The method of claim 8 , further comprising analyzing the video data to identify movement in the detection zone.
12 . The method of claim 8 , further comprising training a model for knock detection based on at least one of historical audio or visual data.
13 . The method of claim 8 , further comprising distinguish between human and non-human presence in the detection zone based on at least one of the audio data or video data.
14 . The method of claim 8 , wherein the notification facilitates interaction of the homeowner to the detection zone.
15 . A system of providing notifications to a user of an alarm system, the system comprising:
a non-transitory memory;
an input/output (I/O) unit;
one or more processors in communication with the memory and I/O unit, the one or more processors being configured to:
capture video data from a detection zone in front of a door;
capture audio data from in front of the door;
detect a knock based on the audio data; and
generate a notification based on the detected knock; and
transmit the notification to a homeowner indicative of the knock.
16 . The system of claim 15 , wherein the processor is further configured to activate a microphone to capture the audio data upon detection of a person based on the video data.
17 . The system of claim 15 , wherein the audio data is captured by a microphone of a doorbell positioned exterior to a front door.
18 . The system of claim 15 , wherein the processor is further configured to analyze the video data to identify movement in the detection zone.
19 . The system of claim 15 , wherein the processor is further configured to train a model for knock detection based on at least one of historical audio or visual data.
20 . The system of claim 15 , wherein the processor is further configured to distinguish between human and non-human presence in the detection zone based on at least one of the audio data or video data.