9 results about "Video based" patented technology

The invention relates to the field of intelligent traffic, in particular to a traffic signal lamp positioning, recognizing and classifying method based on the video analysis. The color change regularity of traffic signal lamps is utilized, in other words, all traffic signal lamps are turned on and turned off regularly and last for a period of time, the color change rule of related pixel points in an image is tracked, the area position of the signal lamps in a surveillance video is recognized, then the position of the signal lamps is further confirmed, and the signal lamps are classified according to the image characteristics of the signal lamps. The method has the advantages that the operation is convenient, the signal lamps can be automatically detected and classified, and the method has good robustness and adaptability and provides the support for the automatic detection of the traffic law violation and signal lamp fault.

The present invention relates generally to both a system and method for visually (e.g., via a video-based system or some other visual system) monitoring one or more objects where such objects are obscured by a high intensity light source such as a plasma, flame, or welding arc. In one embodiment, a system in accordance with the present invention comprises a digital camera, at least one light emitting diode (LED) light source, and at least one filter. In another embodiment, a system in accordance with the present invention comprises a digital camera, at least one light emitting diode (LED) light source, and at least one filter selected from a notch filter, a neutral density filter, or combinations thereof. Additionally, the system of the present invention can further include software designed to process, interpret and/or collect various data captured by the visual monitoring, or, imaging, system of the present invention.

The invention discloses a video-based vehicle tracking method and device. The method comprises: determining a first image frame for tracking a vehicle in a video and using the first image frame to initialize a model; when processing a current image frame, using a determined searching range in a previous image frame to extract image features within the searching range in the current image frame, generates a feature cycle matrix, matching the feature cycle matrix with the current model, and obtaining a matching degree matrix; updating a target position according to a matching degree value of each position of the matching degree matrix to track the vehicle; before generating the model of each image frame, redefining the target size based on the calibration, and zooming the searching range of the previous image frame according to the proportional relation between the target size and the searching range; extracting image features in the searching range in the current image frame, finally generating a model according to the image features, and updating the current model. As the target size of each image frame is updated according to the calibration, the system can track the target in adaptive scales, so that the vehicle tracking is more accurate.

The invention provides a pedestrian real-time detection method based on video. At the detection stage, a calibrated binocular camera is used for acquiring the video, and a disparity map of each frame of image is calculated; each image is divided into a background region and a non-background region through the disparity map; a candidate region, needing detection, of a current frame is determined through a previous-frame detection result, a region which is the non-background region and the candidate region at the same time serves as a region needing detection, and only the region needing detection is detected in the following process. The characteristics of pedestrians can be rapidly and accurately detected, compared with a traditional HOG and other modes, the method is lower in complexity and higher in speed, and can meet real-time and accurate detection requirements.

The embodiment of the invention provides a video frame image testing method, device and equipment and a computer storage medium, and the method comprises the steps: obtaining a to-be-tested video anda reference video, and enabling the version information of the to-be-tested video to be different from the version information of the reference video; obtaining a logic frame number of each test frameimage in the to-be-tested video, and obtaining a logic frame number of each reference frame image in the reference video; respectively determining the similarity between the test frame image and thereference frame image with the same logic frame number to obtain the similarity value of each image; determining a test result of the to-be-tested video based on each image similarity value; and outputting the test result. According to the invention, the completely corresponding test frame image and reference frame image are obtained by using the logic frame number, so that the test accuracy of the video frame image is improved.

The invention discloses a delayed photography method and device, electronic equipment and a storage medium, and belongs to the technical field of visual imaging. The delayed photography method comprises the following steps: when time-delay photography is started and a dynamic vision sensor detects that photography equipment is in a static state, selecting a preparation frame without foreign matters in a preset interval of a left neighborhood or a right neighborhood of a default selection frame by utilizing the dynamic vision sensor; under the condition that the foreign matter exists in the default selection frame, writing the preparation frame into a delay video file; and under the condition that the time-lapse photography is finished, generating a time-lapse photography video based on a time-lapse video file when the time-lapse photography is finished, wherein the default selection frame is a frame which is selected by adopting a fixed time frame selection mode and is used for writing the delay video file.