Human detection with imaging sensors
View Patent ↗A method and apparatus for determining human beings from terrain or man-made obstacles is provided. A long-wave infrared camera along with additional devices such as a color camera, two cameras in stereo configuration, and/or a LADAR scanner are used such that the physical scene captured in one image is the same from all of the devices. The images may be processed such that areas of interest representing characteristics of human beings are labeled likely human. The processing may include determining the physical size, range, and relative locations of the objects found in the images. The system, method and apparatus may be used in unmanned vehicles or autonomous machines, as a driving aid to a manned vehicle or system, or as part of a security system.
1. A method comprising:
capturing an image in a field of view with an infrared camera;
capturing the image in the field of view with another device comprising a stereo camera, wherein output from the stereo camera includes a disparity map;
determining one or more areas of the image that fall within a temperature range indicative of thermal emission characteristics of a human being based on output from the infrared camera;
generating a masked disparity map that includes the disparity map masked with the one or more areas of the image; and
determining that the image includes a human being by determining that a surface area of the masked disparity map is consistent with a surface area of the human being.
2. The method of claim 1 , further including displaying a likelihood that the image includes a human being.
3. The method of claim 1 , further comprising capturing the image in the field of view with a color camera.
4. The method of claim 1 , wherein the infrared camera comprises a long-wave infrared camera.
5. The method of claim 3 , further comprising generating a masked color map that includes an output of the color camera masked with the one or more areas of the image.
6. The method of claim 5 , further comprising determining if the masked color map includes a human face using a face detection algorithm.
7. The method of claim 5 , further comprising determining if the masked color map includes colors that represent human skin colors using a color matching algorithm.
8. An apparatus comprising:
one or more communication interfaces configured to receive data from a field of view simultaneously recorded by an infrared camera and a stereo camera; and
a computer configured to:
determine one or more areas of the field of view that exhibits thermal emission characteristics of a human being based on the data recorded by the infrared camera;
generate a masked disparity map that includes disparity data within the determined one or more areas of the field of view, wherein the disparity data is derived from the data recorded by the stereo camera; and
determine that the field of view includes a human being by determining that a surface area of the masked disparity map is consistent with a surface area of the human being.
9. The apparatus of claim 8 , further comprising a long-wave infrared camera.
10. The apparatus of claim 8 , further comprising the infrared camera and the stereo camera, wherein the infrared and stereo cameras are bore-sighted, such that a physical scene captured in the field of view by each camera is the same.
11. The apparatus of claim 8 , further comprising a color camera.
12. The apparatus of claim 8 , further comprising a Laser-Radar scanner.
13. The method of claim 1 , wherein the temperature range indicative of thermal emission characteristics of the human being is approximately 70 degrees Fahrenheit to 100 degrees Fahrenheit.
14. The method of claim 1 , wherein the determining that the image includes the human being further comprises determining that a volumetric size of the masked disparity map is consistent with a volume of the human being.
15. The apparatus of claim 8 , wherein the computer is further configured to determine that the field of view includes the human being by determining that a volumetric size of the masked disparity map is consistent with a volume of the human being.
16. A non-transitory computer readable memory comprising software, that when executed by a processor, causes the processor to:
receive data from a field of view simultaneously recorded by an infrared camera and a stereo camera;
determine one or more areas of the field of view that exhibits thermal emission characteristics of a human being based on the data recorded by the infrared camera;
generate a masked disparity map that includes disparity data within the determined one or more areas of the field of view, wherein the disparity data is derived from the data recorded by the stereo camera; and
determine that the field of view includes a human being by determining that a surface area of the masked disparity map is consistent with a surface area of the human being.
17. The non-transitory computer readable memory of claim 16 , wherein the software, when executed by the processor, further causes the processor to determine that the image includes the human being by determining that a volumetric size of the masked disparity map is consistent with a volume of the human being.