Abstract: A method and device for fisheye camera automatic calibration is disclosed. A first straight line and a second straight line are respectively moved to generate a first fisheye line and a second fisheye line being straight lines. At least two symmetric points are generated on the second straight line based on an intersection of the first straight line and the second straight line as a symmetric central point, so as to correspondingly generate at least two fisheye symmetric points. The second straight line rotates until a distance between the first fisheye line and each of two fisheye symmetric points is equal. Then, central intersection coordinates, fisheye symmetric coordinates, a fisheye symmetric distance are retrieved and used to calculate fisheye parameters.

Abstract: An automatic cleaning system and an operation method thereof are provided. The automatic cleaning system includes a monitor sensor, a clean robot and a controller. The monitor sensor is disposed on a field and used for monitoring the field to obtain an image. The controller is coupled to the monitor sensor to receive the image, used for analyzing the image to find out an action hot area, and instructs the clean robot to clean the action hot area.

Abstract: An enhanced object detection method uses image discontinuousness for enhancing performance of identifying objects of a specific class in an image data. The method includes retrieving an image data; computing an image discontinuousness value between a first area and other areas surrounding of the first area, which is with different sizes and in different positions within the image data, and marking areas with an image discontinuousness value larger than a threshold; and identifying the objects of the specific class within the sliding window and outputting detection result.

Abstract: An enhanced object detection method uses image discontinuousness for enhancing performance of identifying objects of a specific class in an image data. The method includes retrieving an image data; computing the an image discontinuousness value between a first area and other areas surrounding of the first area, which is with different sizes and in different positions within the image data, and marking areas with an image discontinuousness value larger than a threshold; and identifying the objects of the specific class within the sliding window and outputting detection result.