Abstract: Provided are a method and a device for detecting a flying spot on an edge of a depth image and an electronic device. The method includes the following. Image data of the depth image is collected. A normal vector of each pixel of the depth image is calculated according to the image data. An edge point of the depth image is calculated according to the normal vector with an edge detection operator. The edge point is taken as a seed point, and whether the seed point satisfies a growing condition is determined. Upon determining that the seed point satisfies the growing condition, the seed point is determined as the flying spot.

Abstract: Provided is an apparatus for processing image blurring configured to extract a sample image by calculating a blur generation area through scales of a horizontal length and a vertical length of an original image, read a pixel value corresponding to the blur generation area calculated from the original image and perform a bitwise operation on the sample image, and set, at a pixel of the sample image changed by the bitwise operation, a result value of the bitwise operation. In this case, the bitwise operation performed on the sample image includes applying the same blurring value on n pixel regions adjacent to a corresponding pixel of the sample image to reduce the time for removing image noise.

Abstract: Optical surface degradation detection and remediation systems, devices, and methods are provided herein. An example method includes tracking a first object within images of an area of interest obtained over a period of time by an image sensor of a vehicle during vehicle movement; tracking a light source within the area of interest; determining a change in relative reflective intensity of the first object using the images, based on a light source angle formed between the light source and an optical plane of an optical surface the vehicle, the change in relative reflective intensity being indicative of veiling glare of the optical surface; and activating a remediating measure in response to the veiling glare.

Abstract: A congestion degree estimating unit (12) which estimates, for each image frame, a congestion degree of an area in which each object detected by an object detecting unit (11) is present is provided. An object selecting unit (13) refers to an estimation result of the congestion degree by the congestion degree estimating unit (12) for each image frame, and selects an object present in an area the congestion degree of which is equal to or lower than a first threshold out of one or more objects detected by the object detecting unit (11).

Abstract: An object recognition apparatus is provided with: a recognizer configured to recognize a plurality of objects; an acquirer configured to obtain behavior information; a parallel moving determinator configured to determine whether or not a parallel moving condition is satisfied, indicating that there are a first object and a second object, which move in parallel; a recognition determinator configured to determine whether or not a recognition condition is satisfied, indicating that there is the second object among the objects recognized in second timing; a recognition number determinator configured to determine whether or not a recognition number condition is satisfied, indicating that a number of the objects recognized in the second timing is reduced; and a parallel moving loss determinator configured to determine that the first object is in a parallel-moving-object-lost state, if the parallel moving condition, the recognition condition and the recognition number conditions are satisfied.

Abstract: A method for recognizing hand gestures when the hand is held in front of the user's upper body. The method detects dominant features associated with the upper body, including features on the face (eyes and mouth etc.) and the shoulder extremes. Geometrical relations among these features are used to judge whether the face/head has turned to the side relative to the shoulder. If not, a face-shoulder coordinate system is established using these face-shoulder features, and the hand images are transformed into this new coordinate system. This in effect defines hand gestures relative to the user's face and shoulder, thus significantly reducing the dimensions of the feature space for hand gestures. This face-shoulder coordinate transform can be combined either with traditional appearance-based hand gesture recognition methods, or with a conventional or hybrid deep neural network for RGB or RGB-D images.

Abstract: A method for securing a bonding product in a working region of a bonder via a clamping device. The bonding product and the clamping device are positioned in the working region of the bonder and a partial characteristic contour is captured to determine the position of the bonding product in the working region. The previously set clamping position of the clamping fingers is captured via a camera and a clamping position is calculated. A current position and orientation of clamping fingers and a new misalignment of the clamping fingers is calculated and displayed until the current position and orientation of the clamping fingers corresponds to a clamping finger reference position.