Abstract:
An object identification method is applied to a surveillance system. The surveillance system includes at least one surveillance apparatus. The object identification method includes acquiring a plurality of first feature vectors of a first moving object and a plurality of second feature vectors of at least one second moving object within a series of surveillance images from the surveillance apparatus, transforming the first feature vectors and the second feature vectors respectively into a first cluster distribution set and at least one second cluster distribution set, comparing similarity of the first cluster distribution set and the at least one second cluster distribution set, and setting a similarity ranking of the first cluster distribution set and the at least one second cluster distribution set according to a comparison result so as to determine whether the first moving object and the at least one second moving object are the same.
Abstract:
A surveillance region identifying method is used to analyze a region feature of a surveillance region covered by a surveillance apparatus. The surveillance region identifying method includes analyzing all track information within a series of images acquired by the surveillance apparatus to acquire an appearing point and a disappearing point of each track information, utilizing cluster analysis to define a main appearing point cluster of the appearing points, computing enter vectors of all appearing points inside the main appearing point cluster, and analyzing vector angles of all enter vectors of the main appearing point cluster to define an entrance of the surveillance region.
Abstract:
A motion detection method is applied to a monitoring camera apparatus with motion detection function. The motion detection method includes analyzing a pixel value of each frame from a video stream changed over time, defining a first period and a second period having the pixel value greater than a triggering threshold respectively as a first event and a second event, comparing a time length of the first period with a filtering threshold to acquire time difference between an end point of the first period and a beginning point of the second period, comparing the time difference with a merging threshold, and acquiring relation between the first event and the second event according to comparison results of the filtering threshold and the merging threshold, so as to determine a detecting period of the motion detection function for actuation.
Abstract:
A convolutional neutral network identification efficiency increasing method is applied to a related device. The convolutional neutral network identification efficiency increasing method includes analyzing an input image to acquire foreground information, utilizing the foreground information to generate a foreground mask, and transforming the input image into an output image via the foreground mask. The output image is used to be an input of the convolutional neutral network identification for preferred object identification efficiency.
Abstract:
A method for tagging an object in a video includes playing a video with a plurality of frames, selecting a target object in a playing frame by a cursor, obtaining at least one timestamp and at least one bounding box that correspond to the target object, from an object meta data, showing a selectable area in the playing frame according to the bounding box corresponding to the timestamp of the playing frame, generating at least one tag function item linking to the selectable area, and tagging the target object according to one of the at least one tag function item. Therefore, the target object in the video can be tagged in an easy and fast way.
Abstract:
A surveillance system and method are provided. The surveillance system includes a video capturing device and a processing device. The video capturing device obtains a video and generates metadata associated therewith. The metadata records a trajectory of an object appearing in the video. The processing device executes the video playback method, namely to obtain the metadata, define a section on the trajectory based on information of the trajectory, command a player to play the video by a first speed when the object appears on the section, and selectively command the player to play the video by a second speed when the object appears on the trajectory except the section.
Abstract:
An automatic range finding method is applied to measure a distance between a stereo camera and a reference plane. The automatic range finding method includes 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, and applying a coarse-to-fine computation for the disparity-map video.
Abstract:
An automatic range finding method is applied to measure a distance between a stereo camera and a reference plane. The automatic range finding method includes 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, and applying a coarse-to-fine computation for the disparity-map video.
Abstract:
An object counting method is applied to a surveillance camera and used to determine an amount and a direction an object passing over a surveillance region. The object counting method includes acquiring a plurality of known traces, dividing the plurality of known traces into a first group passing the surveillance region along a first direction and a second group passing the surveillance region along a second direction, computing a first start point computed value of the first group and a second start point computed value of the second group to acquire a start point connection vector, and comparing the start point connection vector with a trace vector of a target object to determine whether the target object passes over the surveillance region along the first direction or the second direction.
Abstract:
An image identifying method is applied to a monitoring camera and a monitoring camera system and used to determine whether a target object is a leaving object or a missing object. The image identifying method includes acquiring a foreground region within a monitoring image corresponding to the target object, analyzing whether the target object inside the foreground region conforms to a variant feature, and comparing the foreground region with a reference image for determining the target object belongs to the leaving object or the missing object when the foreground region does not conform to the variant feature.