1. A miniature autonomous apparatus for scene interpretation, comprising:
a digital camera for producing an image of a scene; and a processor associated with the camera, the processor configured to control a dynamic range control process and an image detection process, wherein the processor is configured to:
acquire a plurality of images of said scene from said digital camera; based on said plurality of images, construct a parametric image and determine an initial parametric representation of each pixel of the parametric image, wherein a new parametric representation is repeatedly calculated for each pixel in said parametric image, based on a current parametric representation of said pixel and a pixel value of a corresponding pixel of an incoming image;
wherein the image detection process includes at least a first process for comparing
pixels of the parametric image with pixels of an incoming image so as to assign for each received pixel a pixel type out of a plurality of possible pixel types including hot pixels and other pixels; compute a binary hot pixels image;
and define at least a first target from the hot pixels; and
wherein at least hot pixels are further processed according to at least a second process not performed upon the other pixels.
2. The apparatus of claim 1, wherein the processor is further configured to:
measure predefined parameters for the first target; and
determine whether the first target is of interest based on application-specific criteria.
3. The apparatus of claim 1, wherein defining at least a first target includes defining at least a first target from the hot pixels based on relative geometric relationships of a plurality of hot pixels.
4. The apparatus of claim 1, wherein the processor is configured to compute an average pixel image and a standard deviation pixel image from the plurality of acquired images.
5. The apparatus of claim 1, wherein the processor is configured to compute a minimum pixel value image and a maximum pixel value image from the plurality of acquired images.
6. The apparatus of claim 1, wherein the processor is configured to compute an average derivative value image and a standard deviation derivative pixel value image from the plurality of acquired images.
7. The apparatus of claim 1, wherein the processor is configured to compute, for each pixel of the parametric representation, at least two of:
a new average pixel value and a new standard deviation value, using the value of a newly acquired pixel and a predetermined weight coefficient;
a new minimum pixel value and a new maximum pixel value, according to the value of a newly acquired pixel; and
a new average derivative pixel value and a new standard deviation derivative value, using the value of a newly acquired pixel and a predetermined weight coefficient.
8. The apparatus of claim 1, wherein the processor is configured to segment hot pixels into one or more connected components.
9. The apparatus of claim 1, wherein the processor is configured to compute geometric centers of gravity of a target in a first image and a previously captured image.
10. The apparatus of claim 6, wherein the processor is configured to compute whether a first pixel is a hot pixel by comparing the difference between an actual derivative value and an average derivative value of the first pixel with the standard deviation derivative of the first pixel.
11. A miniature autonomous apparatus for scene interpretation, comprising:
a digital camera for producing an image of a scene; and
a processor associated with the camera, the processor adapted to control dynamic range and image detection, wherein when dynamic range is changed, an image detection process adapts to the new dynamic range setting, wherein the image detection process includes at least a first process for assigning for each received pixel a pixel type out of a plurality of possible pixel types including hot pixels and other pixels, and wherein at least hot pixels are further processed according to at least a second process not performed upon the other pixels,
wherein the processor is configured to:
determine an initial parametric representation of each pixel of the image;
update the parametric representations according to predefined criteria;
analyze pixels of the image so as to determine which of the pixels are hot pixels;
define at least one target from the hot pixels based on relative geometric relationships of a plurality of hot pixels, wherein defining at least a target includes defining at least a first target from the hot pixels based on relative geometric relationships of a plurality of hot pixels;
compute a new average pixel value and a new standard deviation value, using the value of a newly acquired pixel and a predetermined weight coefficient, a new minimum pixel value and a new maximum pixel value, according to the value of a newly acquired pixel, and a new average derivative pixel value and a new standard deviation derivative value, using the value of a newly acquired pixel and a predetermined weight coefficient;
compute whether a pixel is a hot pixel by comparing the difference between an actual derivative value and an average derivative value of the pixel with the standard deviation derivative of the pixel; and
compute geometric centers of gravity of a target in a first image and a previously captured image.
12. A method for scene interpretation, comprising:
using a processor to produce an image of a scene;
controlling a dynamic range process and an image detection process;
acquire a plurality of images of said scene from said digital camera; based on said plurality of images, construct a parametric image and determine an initial parametric representation of each pixel of the parametric image, wherein a new parametric representation is repeatedly calculated for each pixel in said parametric image, based on a current parametric representation of said pixel and a pixel value of a corresponding pixel of an incoming image;
wherein the image detection process includes at least a first process for
comparing pixels of the parametric image with pixels of an incoming image so as to assign for each received pixel a pixel type out of a plurality of possible pixel types including hot pixels and other pixels; and compute a binary hot pixels image;
defining at least one target from the hot pixels based on relative geometric relationships of a plurality of hot pixels; and
wherein at least hot pixels are further processed according to at least a second process not performed upon the other pixels.
13. The method of claim 12, further comprising computing an average pixel image and a standard deviation pixel image from the plurality of acquired images.
14. The method of claim 12, further comprising computing a minimum pixel value image and a maximum pixel value image from the plurality of acquired images.
15. The method of claim 12, further comprising computing an average derivative value image and a standard deviation derivative pixel value image from the plurality of acquired images.
16. The method of claim 14, further comprising computing an average derivative value image and a standard deviation derivative pixel value image from the plurality of acquired images.
17. The method of claim 12, further comprising segmenting hot pixels into one or more connected components.
18. The method of claim 12, further comprising determining whether a first pixel is a hot pixel by comparing the difference between an actual derivative value and an average derivative value of the pixel with the standard deviation derivative of the first pixel.
19. The method of claim 12, further comprising determining geometric centers of gravity of a target in a first image and a previously captured image.
The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.
1. A furnace for burning wood particles, comprising:
an outer layer, having a substantially cylindrical shape;
an inner layer, having a substantially cylindrical shape, disposed inside the outer layer and together forming a substantially cylindrically shaped pre-heating non-combustion and non-exhaust zone for pre-combustion air to pass before entering a combustion zone which is interior of the inner layer, through a longitudinally intermediate side port sized, located and configured to cause pre-combustion air to flow into said combustion zone interior of the inner layer along a line which is substantial tangential to said inner layer and then flow within said combustion zone interior to the inner layer in a substantially circular path;
a bottom disposed under the inner layer and a top disposed above the inner layer;
a bottom fed particulate fuel supply means configured to feed particulate fuel through a void in the bottom and further configured to automatically limit a height characteristic of fuel disposed in said combustion zone interior of the inner layer and above the bottom; and
an exhaust means configured to transport heated air, unburned fuel and ash from inside of to outside of the inner layer, further comprising a centrally disposed heat exhaust tube extending downwardly into a middle section of the interior of the inner layer.
2. A furnace of claim 1 further comprising an opening in the inner layer configured to aiding in lighting fuel within the inner layer.
3. The furnace of claim 2 further comprising an inner door on the opening in the inner layer.
4. The furnace of claim 3 further comprising an outer door on an opening in the outer layer, in registration with the inner door.
5. The furnace of claim 4 wherein the fuel level limit controller comprises a pivoting arm.
6. A furnace of claim 1 wherein the bottom fed particulate fuel supply means comprises an auger in a pipe.
7. A furnace of claim 1 further comprising a blower for forcing air into the interior of the inner layer.
8. The furnace of claim 1 wherein the pre-heating zone extends from the bottom up to a top of the inner layer.
9. The furnace of claim 8 wherein the particulate fuel comprises wood chips.
10. The furnace of claim 1 wherein the outer layer is made of steel.
11. The furnace of claim 10 wherein the inner layer is made of steel.
12. A furnace for burning wood particles, comprising:
an outer enclosing member having a substantially open interior;
an inner enclosing member, having a substantially open interior, the inner enclosing member disposed co-axially inside the outer enclosing member and together forming an interior limit and an exterior limit of a non-combustion, non-exhaust, preheating zone for pre-combustion air to pass before entering a combustion chamber interior of the inner enclosing member;
a bottom fed particulate fuel supply means configured to feed particulate fuel upwardly and further configured to automatically limit a height characteristic of fuel disposed in the combustion chamber;
an exhaust means configured to transport heated air, unburned fuel and ash from the combustion chamber, further comprising a central disposed heat exhaust tube extending downwardly into a middle section of the interior of the inner enclosing member; and
wherein said inner enclosing member is configured as a cylinder with a longitudinally intermediate interior air inlet port therein for causing said combustion air to flow in a substantially circular part around, within and adjacent to said inner enclosing member.
13. A furnace of claim 12 wherein the combustion chamber is cylindrical and co-axially disposed within the outer enclosing member.
14. A furnace of claim 12 further comprising a bottom disposed beneath and coupled to said inner enclosing member and said outer enclosing member.
15. A furnace of claim 12 wherein said outer enclosing member is configured as a cylinder and further comprising an exterior air inlet port therein.
16. A furnace of claim 12 wherein said bottom fed particulate fuel supply means is configured to sense a height characteristic of fuel disposed within the combustion chamber and to limit further introduction of fuel into the combustion zone when the height characteristic exceeds a predetermined threshold.
17. A furnace of claim 12 further comprising a means for facilitating air to be blown into the pre-heating zone.