IP security camera combining both infrared and visible light illumination plus sensor fusion to achieve color imaging in zero and low light situations
An apparatus includes a first light source, a second light source, an image sensor circuit, and a processing circuit. The first light source is generally capable of emitting infrared light. The second light source is generally capable of emitting visible light. The image sensor circuit is generally responsive to both the infrared light and the visible light. The processing circuit is generally coupled to the image sensor circuit and configured to generate an image comprising both infrared information and color information.
1 . An apparatus comprising:
a first light source capable of emitting infrared light to illuminate a scene;
a second light source capable of emitting visible light to illuminate said scene;
an RGB-IR image sensor circuit comprising an RGB-IR image sensor responsive to both infrared and visible light, and configured to generate video surveillance data of said scene comprising both infrared information and color information, wherein said RGB-IR image sensor being responsive to said infrared light emitted by said first light source reduces an amount of visible light illumination needed to obtain said color information; and
a processing circuit coupled to said first light source, said second light source, and said RGB-IR image sensor circuit, wherein (a) said processing circuit includes a video processing pipeline comprising a directional selection circuit and (b) said processing circuit is configured to (i) automatically control said first light source, said second light source, and said RGB-IR image sensor circuit, (ii) process said video surveillance data arranged as video frames, (iii) generate a composite image comprising both the infrared information and the color information contained in said video surveillance data received from said RGB-IR image sensor circuit, (iv) utilize said directional selection circuit of said video processing pipeline to analyze signals comprising at least audio signals generated from sounds picked up in a field of view monitored by the RGB-IR image sensor by one or more sensors coupled to said processing circuit to determine directional information of an object in the field of view of said RGB-IR image sensor, (v) utilize said directional information to identify a corresponding location of an area of interest in said video frames, and (vi) apply a person detection operation to said infrared information in said area of interest utilizing a neural network trained to separate people from other objects to identify a source of the sounds.
2 . The apparatus according to claim 1 , wherein said processing circuit is further configured to control turn on of said first light source and said second light source.
3 . The apparatus according to claim 1 , wherein:
said first light source comprises one or more infrared light emitting diodes (LEDs); and
said second light source comprises one or more white light LEDs.
4 . The apparatus according to claim 1 , wherein said apparatus is configured as at least one of a doorbell camera, an exterior security camera, and an interior security camera.
5 . The apparatus according to claim 1 , wherein said apparatus is configured as one or more of an interior monitoring camera configured to monitor a vehicle passenger compartment, an external short range camera configured to provide around vehicle monitoring, an automatic parking assistance camera, and a rear view camera.
6 . The apparatus according to claim 1 , wherein said processing circuit comprises a camera system on chip.
7 . The apparatus according to claim 1 , further comprising a motion detection circuit coupled to said processing circuit and configured to activate said processing circuit in response to detecting motion in a predefined field of view.
8 . The apparatus according to claim 7 , wherein said motion detection circuit comprises a passive infrared (PIR) sensor.
9 . The apparatus according to claim 1 , wherein said apparatus is configured as a battery-powered camera.
10 . The apparatus according to claim 1 , wherein said processing circuit is configured to perform a sensor fusion operation utilizing infrared image data and said color information captured using said RGB-IR image sensor.
11 . The apparatus according to claim 1 , wherein said video processing pipeline of said processing circuit further comprises a facial analysis module configured to determine whether a face detected in said area of interest is a person matching one of a plurality of user profiles stored in a database of known faces.
12 . The apparatus according to claim 11 , wherein said processing circuit is further configured to determine a distance between said object and said RGB-IR image sensor by applying one or more image processing operations on said video surveillance data from said RGB-IR image sensor.
13 . The apparatus according to claim 11 , wherein said processing circuit is further configured to turn on said first light source in response to detection of said object in said field of view of said RGB-IR image sensor and determine whether said object detected in said field of view of said RGB-IR image sensor is closer than a predefined distance from said RGB-IR image sensor.
14 . The apparatus according to claim 13 , wherein said processing circuit is further configured to turn on said second light source in response to said object in said field of view of said RGB-IR image sensor being closer than said predefined distance from said RGB-IR image sensor.
15 . The apparatus according to claim 14 , wherein said processing circuit is further configured to turn on said second light source to illuminate at least one of said person, a key pad, a touch pad, and a touch screen.
16 . The apparatus according to claim 14 , wherein said processing circuit is further configured to (i) perform video operations to increase resolution and zoom in on the area of interest, (ii) crop out a portion of the area of interest containing said face of said person, (iii) generate a color image of said person, and (iv) perform face recognition operations on said color image.
17 . The apparatus according to claim 16 , wherein said processing circuit is further configured to determine access privileges of said person based upon said face recognition operations on said color image.
18 . A method of color imaging in zero and low light situations comprising the steps of:
detecting motion in a field of view of a security camera using one or more sensors, wherein said security camera comprises (i) a first light source capable of emitting infrared light to illuminate a scene, (ii) a second light source capable of emitting visible light to illuminate said scene, (iii) an RGB-IR image sensor circuit comprising an RGB-IR image sensor responsive to both infrared and visible light, and configured to generate video surveillance data comprising both infrared information and color information, and (iv) a processing circuit coupled to said first light source, said second light source, and said RGB-IR image sensor circuit, and configured to control said first light source, said second light source, and said RGB-IR image sensor circuit;
automatically turning on said first light source capable of emitting infrared light and said RGB-IR image sensor circuit in response to detection of said motion in said field of view of said security camera, wherein said processing circuit is configured to (i) process said video surveillance data of said scene arranged as video frames, (ii) generate a composite image comprising both said infrared information and said color information, (iii) utilize a directional selection circuit of a video processing pipeline of said processing circuit to analyze signals comprising at least audio signals generated from sounds picked up in a field of view monitored by the RGB-IR image sensor by one or more sensors coupled to said processing circuit to determine directional information of an object in the field of view of said security camera, (iv) utilize said directional information to identify a corresponding location of an area of interest in said video frames, and (v) apply a person detection operation to said infrared information in said area of interest utilizing a neural network trained to separate people from other objects to identify a source of the sounds;
determining a distance between said object in said field of view of said security camera by applying one or more image processing operations on said video surveillance data using said processing circuit coupled to said RGB-IR image sensor circuit; and
when said distance is less than a predefined distance, turning on said second light source capable of emitting visible light and using said RGB-IR image sensor circuit and said processing circuit to generate a color image comprising both said infrared information and said color information, wherein said RGB-IR image sensor being responsive to said infrared light emitted by said first light source reduces an amount of visible light illumination needed to obtain said color information.
19 . The method according to claim 18 , wherein:
when said object is determined to be a person, (i) performing video operations to increase resolution and zoom in on the area of interest, (ii) cropping out a portion of the area of interest of said color image containing a face of said person, and (iii) performing face recognition operations on said cropped out portion of said color image.
20 . The method according to claim 19 , further comprising determining access privileges of said person based upon said face recognition operations on said color image.