Stereo camera and automatic range finding method for measuring a distance between stereo camera and reference plane

1. An automatic range finding method capable of measuring a distance between a stereo camera and a reference plane, the automatic range finding method comprising:

acquiring a disparity-map video by the stereo camera facing the reference plane;

analyzing the disparity-map video to generate a depth histogram;

selecting a pixel group having an amount greater than a threshold from the depth histogram;

calculating the distance between the stereo camera and the reference plane by weight transformation of the pixel group;

acquiring at least one parameter of an object according to the distance between the stereo camera and the reference plane;

utilizing field of view information of the stereo camera and the distance between the stereo camera and the reference plane to decide tracing computing parameters M and N, wherein M and N are positive integers;

dividing the disparity-map video into a plurality of coarse candidate zones wherein each coarse candidate zone has a length of M pixels and a width of N pixels;

determining whether to set at least one coarse selected zone from the plurality of coarse candidate zones;

enlarging the coarse selected zone to acquire a coarse selected enlarged zone;

dividing the coarse selected enlarged zone into a plurality of fine candidate zones wherein each fine candidate zone has a length of m pixels and a width of n pixels, wherein m and n are positive integers;

determining at least one fine selected zone from the plurality of fine candidate zones; and

acquiring position of a mass center of the at least one fine selected zone.

2. The automatic range finding method of claim 1 , wherein the at least one parameter is selected from a group consisting of a minimal distance between the stereo camera and the object and a size of the object.

3. The automatic range finding method of claim 1 , wherein a step of acquiring the at least one parameter of the object comprises:

utilizing a mapping table to acquire the at least one parameter of the object, or utilizing the mapping table and interpolation calculation to acquire the at least one parameter of the object.

4. The automatic range finding method of claim 1 , wherein a step of determining the at least one fine selected zone from the plurality of fine candidate zones comprises:

calculating an integral image value of the disparity-map video by an integral image computing method;

calculating a pixel average value of the coarse selected enlarged zone according to the integral image value;

comparing a pixel average value of the fine candidate zone with a pixel average value of the coarse selected enlarged zone; and

setting the fine candidate zone having a pixel average value greater than or equal to the pixel average value of the coarse selected enlarged zone as the at least one fine selected zone.

5. The automatic range finding method of claim 1 , further comprising:

selecting a plurality of verifying zones around the mass center;

comparing a pixel value of the mass center with a pixel average value of each verifying zone; and

determining whether to execute tracing computation according to a comparison result.

6. The automatic range finding method of claim 5 , wherein the position of the mass center is utilized to execute the tracing computation while an amount of the verifying zone having the pixel average value smaller than or equal to the pixel value of the mass center is greater than or equal to a specific amount.

7. A stereo camera with an automatic range finding function capable of measuring a distance relative to a reference plane, the stereo camera comprising:

an image sensor facing the reference plane to acquire a disparity-map video containing an object; and

an operating processor electrically connected to the image sensor and adapted to analyze the disparity-map video to generate a depth histogram, select a pixel group having an amount greater than a threshold from the depth histogram and calculate the distance between the stereo camera and the reference plane by weight transformation of the pixel group, so as to acquire at least one parameter of an object according to the distance between the stereo camera and the reference plane, to utilize field of view information of the stereo camera and the distance between the stereo camera and the reference plane to decide tracing computing parameters M and N, to divide the disparity-map video into a plurality of coarse candidate zones wherein each coarse candidate zone has a length of M pixels and a width of N pixels, to determine whether to set at least one coarse selected zone from the plurality of coarse candidate zones, to enlarge the coarse selected zone to acquire a coarse selected enlarged zone, to divide the coarse selected enlarged zone into a plurality of fine candidate zones wherein each fine candidate zone has a length of m pixels and a width of n pixels, to determine at least one fine selected zone from the plurality of fine candidate zones, and to acquire position of a mass center of the at least one fine selected zone, wherein M and N are positive integers, and m and n are positive integers.

8. An automatic range finding method capable of measuring a distance between a stereo camera and a reference plane, the automatic range finding method comprising:

acquiring a disparity-map video by the stereo camera facing the reference plane;

analyzing the disparity-map video to generate a depth histogram;

selecting a pixel group having an amount greater than a threshold from the depth histogram;

calculating the distance between the stereo camera and the reference plane by weight transformation of the pixel group;

acquiring at least one parameter of an object according to the distance between the stereo camera and the reference plane;

utilizing field of view information of the stereo camera and the distance between the stereo camera and the reference plane to decide tracing computing parameters M and N, wherein M and N are positive integers;

dividing the disparity-map video into a plurality of coarse candidate zones wherein each coarse candidate zone has a length of M pixels and a width of N pixels;

calculating an integral image value of the disparity-map video by an integral image computing method;

calculating a pixel average value of each coarse candidate zone according to the integral image value;

comparing the pixel average value of the each coarse candidate zone with a pixel average value of an adjacent coarse candidate zone abutting upon the each coarse candidate zone; and

setting a coarse candidate zone having a pixel average value greater than or equal to the pixel average value of the adjacent coarse candidate zone as a coarse selected zone.

9. An automatic range finding method capable of measuring a distance between a stereo camera and a reference plane, the automatic range finding method comprising:

acquiring a disparity-map video by the stereo camera facing the reference plane;

analyzing the disparity-map video to generate a depth histogram;

selecting a pixel group having an amount greater than a threshold from the depth histogram;

calculating the distance between the stereo camera and the reference plane by weight transformation of the pixel group;

acquiring at least one parameter of an object according to the distance between the stereo camera and the reference plane;

utilizing field of view information of the stereo camera and the distance between the stereo camera and the reference plane to decide tracing computing parameters M and N, wherein M and N are positive integers;

dividing the disparity-map video into a plurality of coarse candidate zones wherein each coarse candidate zone has a length of M pixels and a width of N pixels;

calculating an integral image value of the disparity-map video by an integral image computing method;

calculating a pixel average value of each coarse candidate zone according to the integral image value;

calculating a whole pixel average value of the disparity-map video;

comparing the pixel average value of the each coarse candidate zone with the whole pixel average value of the disparity-map video; and

setting a coarse candidate zone having a pixel average value greater than or equal to the whole pixel average value of the disparity-map video as a coarse selected zone.