33results about How to "Real-time video surveillance" patented technology

The invention discloses a Delta robot-based target tracking method and system. The method comprises the steps of judging whether a person enters a target region or not based on an infrared sensor on a Delta robot; detecting a human body target object in the target region by adopting a region-based method; starting visual tracking, and locating the human body target object by a coplanar P4P-based monocular vision locating principle; after the locating of the human body target object is finished, performing video tracking on the human body target object by a structure similarity-based video tracking method; and transmitting a video obtained by video tracking to a control room in real time, and cautioning a state of the human body target object in video display. Through the method and the system provided by embodiments of the invention, a person object can be effectively detected, so that the target is located, the visual tracking effect in an industrial control process is ensured, and preparations are made for accurate understanding and optimization of a person.

The invention discloses a background subtraction method based on color constant parameters. By defining the color constant parameters directing at RGB (red green blue) images as main basis for distinguishing background pixels from foreground moving targets, the method solves the problem that foreground targets have holes due to color similarity in background subtraction technology. Besides, the method is efficient in running, small in occupied computation memory, short in consumed time, suitable for embedded application and capable of realizing real-time video monitoring.

A system for monitoring dangerous behaviors of escalator passengers based on visual analysis, including a video access module, an intelligent data acquisition module and a cloud platform. The intelligent data acquisition module includes a video acquisition module, an algorithm calculation module, and a video stream storage module; the video stream storage module collects The video stream data of the video access module, and transmit the video stream data to the algorithm calculation module. The algorithm calculation module analyzes and calculates the collected video stream data to obtain the analysis results, and stores the analysis results in the video according to different data contents. Stream storage module and cloud platform. The present invention proposes a system and method for monitoring dangerous behaviors of escalator passengers based on visual analysis, which can accurately identify the behaviors of people and objects on the escalator, and make early warnings to achieve the function of reminder and prevention, and reduce the occurrence of escalator accidents.

The invention relates to an adaptive microphone array sound positioning rescue robot and a using method thereof. A robot comprises a vehicle body having a running mechanism, a main control unit on the vehicle body and connected with a running driving motor, an obstacle avoiding module and a sound positioning module, wherein the obstacle avoidance module is three sets of supersonic wave detectors, supersonic waves are respectively emitted from the right ahead portion, the front side portion and the front upper portion, an electric cylinder push rod end of the sound positioning module fixed on the vehicle body is equipped with a supporting block and a supporting board, the gear shaft penetrates through the supporting board of which upper and lower surfaces are respectively equipped with a gear and two tooth bar guide rods, and an outer end of each tooth bar guide rod is fixedly equipped with a microphone array element to form a cross array. According to the using method, the three sets of supersonic wave detectors are controlled by a main control unit to detect distances between obstacles and the vehicle body, height and array element separation distances of the microphone arrays are adaptively maintained to be the largest at the maximum degree, moreover, sound received by four array elements is analyzed, positioning on a help-seeking sound source is carried out to correct a navigation angle to approach the help-seeking sound source. Through the using method, a complex disaster site can be rapidly and safely reached, sound source positioning precision is high, and searching efficiency is high.

The invention discloses a panorama monitoring system and a method thereof, wherein the panorama monitoring system comprises a monitoring center and a plurality of lenses which are connected with the monitoring center. The panorama monitoring system is characterized in that the monitoring center comprises video collecting equipment, digital picture collecting equipment, region dividing and encoding equipment and video overlapping equipment. The video collecting equipment is configured to receive video data from the plurality of lenses. The digital picture collecting equipment is configured to receive picture data from a digital picture source. The region dividing and encoding equipment is configured to divide chips on a monitoring region and encode the divided chips. The video overlapping equipment is connected with the video collecting equipment and the digital picture collecting equipment and is configured to overlap the video data which correspond to the plurality of lenses within the chip range of the encoding with picture data according to the encoding and generates panorama video data, and therefore real-time overall consistent three-dimensional panoramic video monitoring of a region, a large region and even an extra large region is achieved.