User detection system and image processing device

1. A user detection system, comprising:

a camera which is installed in a peripheral region of a door, and captures a running region where the door runs when opening or closing the door and a region near the door;

a boundary detector which detects a boundary between a groove sill for opening/closing the door as a first structure and a floor as a second structure in the region near the door based on an image captured by the camera;

a user detector which detects whether there is a user in the running region based on a result detected by the boundary detector; and

a controller which controls an open/close operation of the door based on a result detected by the user detector.

2. The user detection system of claim 1 ,

wherein the boundary detector detects a straight boundary between the first structure and the second structure, and

wherein the user detector detects whether there is a user in the running region based on whether the straight boundary detected by the boundary detector is broken.

3. The user detection system of claim 2 ,

wherein the user detector determines that there is a user in the running region in a case where the straight boundary is broken.

4. The user detection system of claim 1 ,

wherein the boundary detector sets a detection area based on a dimension of the door, and narrows an area to detect the boundary down to the detection area.

5. The user detection system of claim 1 ,

wherein the boundary detector detects the boundary based on one image which is captured by the camera.

6. The user detection system of claim 1 ,

wherein the boundary detector binarizes the image captured by the camera to detect the boundary between the first structure and the second structure, and

wherein the user detector

focuses on a pixel group located in an x-th column among a number of pixels of the binarized image,

calculates a total sum of pixel values of the focused pixel group, and

detects whether there is a user in the running region based on whether the calculated total sum is “0”.

7. The user detection system of claim 1 ,

wherein the boundary detector binarizes the image captured by the camera to detect the boundary between the first structure and the second structure, and

wherein the user detector

focuses sequentially on pixel columns arranged in a horizontal direction among a number of pixels which are arranged in the horizontal direction and a vertical direction of the binarized image,

calculates total sums of pixel values of the focused pixel columns sequentially, and

detects whether there is a user in the running region based on whether a ratio of the pixel columns of which the calculated total sum is equal to or more than “1” is less than a threshold.

8. A detection system, comprising:

a camera, positioned in a peripheral region of a door, configured to capture a running region where the door runs when opening or closing the door and a region near the door;

a boundary detector configured to detect a boundary between a tip end of the door as a first structure and an entrance column installed in a region near the door as a second structure in the region near the door based on an image captured by the camera;

an object detector configured to detect whether there is an object in the running region based on a result detected by the boundary detector; and

a controller configured to control an open/close operation of the door based on a result detected by the object detector.

9. The detection system of claim 8 ,

wherein the boundary detector configured to detect a straight boundary between the first structure and the second structure, and

wherein the object detector is configured to detect whether there is an object in the running region based on whether the straight boundary detected by the boundary detector is broken.

10. The detection system of claim 9 ,

wherein the object detector is configured to determine that there is an objectd in the running region in a case where the straight boundary is broken.

11. The detection system of claim 8 ,

wherein the boundary detector is configured to set a detection area based on a dimension of the door, and narrow an area to detect the boundary down to the detection area.

12. The detection system of claim 8 ,

wherein the boundary detector is configured to detect the boundary based on one image which is captured by the camera.

13. The detection system of claim 8 ,

wherein the boundary detector is configured to binarize the image captured by the camera to detect the boundary between the first structure and the second structure, and

wherein the object detector is configured to

focus on a pixel group located in an x-th column among a number of pixels of the binarized image,

calculate a total sum of pixel values of the focused pixel group, and

detect whether there is an object in the running region based on whether the calculated total sum is “0”.

14. The detection system of claim 8 ,

wherein the boundary detector is configured to binarize the image captured by the camera to detect the boundary between the first structure and the second structure, and

wherein the object detector is configured to

focuse sequentially on pixel columns arranged in a horizontal direction among a number of pixels which are arranged in the horizontal direction and a vertical direction of the binarized image,

calculate total sums of pixel values of the focused pixel columns sequentially, and

detect whether there is an object in the running region based on whether a ratio of the pixel columns of which the calculated total sum is equal to or more than “1” is less than a threshold.

15. A detection system, comprising:

a camera, positioned in a peripheral region of a door, and configured to capture a running region where the door runs when opening or closing the door and a region near the door;

a boundary detector configured to detect a boundary between a tip end of the door as a first structure and a floor as a second structure in the region near the door based on an image captured by the camera;

an object detector configured to detect whether there is an object in the running region based on a result detected by the boundary detector; and

a controller which controls an open/close operation of the door based on a result detected by the object detector.

16. The detection system of claim 15 ,

wherein the boundary detector is configured to detect a straight boundary between the first structure and the second structure, and

wherein the user detector s configured to detect whether there is an object in the running region based on whether the straight boundary detected by the boundary detector is broken.

17. The detection system of claim 16 ,

wherein the object detector is configured to determine that there is an object in the running region in a case where the straight boundary is broken.

18. The detection system of claim 15 ,

wherein the boundary detector is configured to set a detection area based on a dimension of the door, and narrow an area to detect the boundary down to the detection area.

19. The detection system of claim 15 ,

wherein the boundary detector is configured to detect the boundary based on one image which is captured by the camera.

20. The detection system of claim 15 ,

wherein the boundary detector is configured to binarize the image captured by the camera to detect the boundary between the first structure and the second structure, and

wherein the object detector is configured to

focus on a pixel group located in an x-th column among a number of pixels of the binarized image,

calculate a total sum of pixel values of the focused pixel group, and

detect whether there is a object in the running region based on whether the calculated total sum is “0”.

21. The detection system of claim 15 ,

wherein the boundary detector is configured to binarize the image captured by the camera to detect the boundary between the first structure and the second structure, and

wherein the object detector is configured to

focus sequentially on pixel columns arranged in a horizontal direction among a number of pixels which are arranged in the horizontal direction and a vertical direction of the binarized image,

calculate total sums of pixel values of the focused pixel columns sequentially, and

detect whether there is an object in the running region based on whether a ratio of the pixel columns of which the calculated total sum is equal to or more than “1” is less than a threshold.