Stereo matching method and system using rectangular window
Disclosed is a stereo matching method and apparatus based on a stereo vision, the method including acquiring a left image and a right image, identifying image data by applying a window to each of the acquired left image and right image, storing the image data in a line buffer, extracting a disparity from the image data stored in the line buffer, and generating a depth map based on the extracted disparity.
1. A stereo matching method based on a stereo vision, the method comprising:
acquiring a left image and a right image;
identifying image data by applying a window to each of the acquired left image and right image;
storing the image data in a line buffer;
extracting a disparity from the image data stored in the line buffer; and
generating a depth map based on the extracted disparity,
wherein the depth map is generated based on the image data identified by applying the window which is a rectangular window that has a same area as a square window and is determined to have a line buffer with a length less than a vertical length of a line buffer formed in the square window.
2. The stereo matching method of claim 1 , wherein, in the rectangular window, a memory usage for storing the left image and the right image is minimized when a vertical length of the line buffer is reduced.
3. The stereo matching method of claim 1 , wherein the line buffer of the rectangular window has a horizontal length that is twice a vertical length.
4. The stereo matching method of claim 1 , wherein the identifying of the image data comprises:
identifying at least one pair of correspondence points at a same position in the acquired left image and right image; and
applying the window based on the at least one pair of correspondence points.
5. The stereo matching method of claim 1 , wherein the extracting of the disparity comprises:
removing a y-axis directional disparity in the acquired left image and right image; and
extracting an x-axis directional disparity in the left image and right image from which the y-axis directional disparity is removed.
6. The stereo matching method of claim 5 , wherein the extracting of the x-axis directional disparity comprises:
segmenting the window into columnar units and performing a chain type comparison operation on pixels included in each of windows segmented into columnar units; and
extracting the x-axis directional disparity by accumulating the segmented windows on which the chain type comparison operation is performed.
7. A stereo matching system based on a stereo vision, the system comprising:
a stereo camera configured to capture a left image and a right image;
a memory configured to store a plurality of line buffers; and
a processor configured to:
acquire the left image and the right image using the stereo camera;
identify image data by applying a window to each of the acquired left and right images;
store the image data in a line buffer;
extract a disparity from the image data stored in the line buffer; and
generate a depth map based on the extracted disparity, and
wherein the depth map is generated based on the image data identified by applying the window which is a rectangular window that has a same area as a square window and is determined to have a line buffer with a length less than a vertical length of a line buffer formed in the square window.
8. The stereo matching system of claim 7 , wherein, in the rectangular window, a memory usage for storing the left image and the right image is minimized when a vertical length of the line buffer is reduced.
9. The stereo matching system of claim 7 , wherein the line buffer of the rectangular window has a horizontal length that is twice a vertical length.
10. The stereo matching system of claim 7 , wherein the processor is configured to identify at least one pair of correspondence points at a same position in the acquired left image and right image, and apply the window based on the at least one pair of correspondence points.
11. The stereo matching system of claim 7 , wherein the processor is configured to remove a y-axis directional disparity in the acquired left image and right image, and extract an x-axis directional disparity in the left image and right image from which the y-axis directional disparity is removed.
12. The stereo matching system of claim 11 , wherein the processor is configured to segment the window into columnar units and perform a chain type comparison operation on pixels included in each of windows segmented into columnar units, and extract the x-axis directional disparity by accumulating the segmented windows on which the chain type comparison operation is performed.
13. The stereo matching system of claim 7 , wherein the processor comprises:
a preprocessor configured to rectify a y-axis directional disparity in the left image and the right images;
a matcher configured to apply a preset size of window to the left image and the right image; and
a postprocessor configured to generate a depth map based on image data output from the matcher.