156 results about "Aerial video" patented technology

In some implementations, a camera may be disposed on an autonomous aerial platform. A user may operate a smart wearable device adapted to configured, and/or operate video data acquisition by the camera. The camera may be configured to produce a time stamp, and/or a video snippet based on receipt of an indication of interest from the user. The aerial platform may comprise a controller configured to navigate a target trajectory space. In some implementation, a data acquisition system may enable the user to obtain video footage of the user performing an action from the platform circling around the user.

The invention discloses a real-time panoramic image stitching method of aerial videos shot by an unmanned plane. The method comprises the steps of: utilizing a video acquisition card to acquire images which are transmitted to a base station in real time by an unmanned plane through microwave channels, carrying out key frame selection on an image sequence, and carrying out image enhancement on key frames; in the image splicing process, firstly carrying out characteristic detection and interframe matching on image frames by adopting an SURF (speeded up robust features) detection method with good robustness; then reducing the series-multiplication accumulative errors of images in a frame-to-mosaic image transformation mode, determining images which are not adjacent in time sequence but adjacent in space on a flight path according to the GPS (global positioning system) position information of the unmanned plane, optimizing the frame-to-mosaic transformation relation, determining image overlapping areas, thereby realizing image fusion and the panoramic image construction and realizing real-time effect of carrying out flying and stitching simultaneously; and in image transformation, based on adjacent frame information in a vision field and adjacent frame information in airspace, optimizing image transformation to obtain the accurate panoramic images. The stitching method has good real-time performance, is fast and accurate and meets the requirements of application occasions in multiple fields.

A method and system are disclosed for creating a real-time, high accuracy mosaic from an aerial video image stream by applying orthorectification of each original video image frame using known ground control points, utilizing a photogrammetric model resolving the object image into pixilation, applying shading to the pixellation, and mosaicking the shaded pixilation of several orthorectified images into a mosaicked image where the mosaicked image is then scaled to the known original image dimensions.

The invention discloses a detection and tracking method of a moving small target in an aerial shot video. The method includes the steps of 1, collecting images, 2, extracting SURF feature points, 3, carrying out grouped matching on the images, 4, obtaining an affine matrix, 5, obtaining a difference image, 6, carrying out opening operation, 7, extracting a target area, 8, determining a target template, 9, determining a target detection area, 10, extracting and matching the feature points, 11, determining a registering center position of a target, 12, determining a target central position, and 13, determining the length and the width of the target. The method has good real-time performance and robustness in the target tracking process, and can obtain smooth target moving tracks in the target tracking process.

The invention provides an automatic lane line identification method based on low-altitude aerial images. The automatic lane line identification method based on the low-altitude aerial images is applied to the field of intelligent transportation. The automatic lane line identification method based on the low-altitude aerial images comprises the steps that (1) an original image of a road is collected through low-altitude aircrafts, and the situation that the shooting angle of the aerial photography road is in the horizontal direction, and the area of the road is arranged in the middle of the image is ensured; (2) the collected image is converted into a grayscale image, the contrast ratio is improved, the grayscale image is copied, and an edge detection image and a binarization image are obtained according to the grayscale image; (3) connected areas are detected in the edge detection image and the binarization image, and characteristics of the connected areas are recorded; (4) the connected areas are deleted according to the number of pixels of the connected areas, the number of connected boundaries, the size of a bounding rectangle and variance values to obtain a central road line; (5) search is carried out on the two sides of the central road line to find roadside lane lines. The automatic lane line identification method based on the low-altitude aerial images is easy and convenient to calculate, and high in arithmetic speed and reliability, the road portion in an aerial photography video can be effectively extracted, straight roads and curve roads in the aerial images can be detected, and the automatic lane line identification method based on the low-altitude aerial images cannot be disturbed by background changes, and is high in accuracy.

In some implementations, a camera may be disposed on an autonomous aerial platform. A user may operate a smart wearable device adapted to configured, and/or operate video data acquisition by the camera. The camera may be configured to produce a time stamp, and/or a video snippet based on receipt of an indication of interest from the user. The aerial platform may comprise a controller configured to navigate a target trajectory space. In some implementation, a data acquisition system may enable the user to obtain video footage of the user performing an action from the platform circling around the user.

The invention discloses an aerial photographic video tracking method based on scale self-adaptive and shielding handling. The method comprises the steps that as for each frame of input image of an aerial photographic video, the peak value of a KCF tracker response function under the current scale is calculated, whether a target is shielded is judged, updating of the tracker parameters is stopped if the judgment result is yes, or scale change detection is continued; and a large scale or a small scale is calculated through a threshold limiting method every time, the optical scale for describing the current target state is detected and the tracker parameters of the current scale are updated. The method has great robustness and real-time performance in the process of tracking the target in the aerial photographic video so that accurate and continuous tracking of the aerial photographic target can be realized.

The invention brings forward a rapid and automatic mosaic technology of aerial video in a search and tracking system. The technology provided by the invention comprises the following steps: two continuous frames of color images in an aerial video sequence are converted to grayscale images; three-valued processing is carried out on the grayscale images by an image three-valued method, edge feature information images are extracted by an edge detection sobel operator and feature points are extracted by an Harris algorithm; feature point matching is completed by an optical flow pyramid algorithm; optimal homography matrix of the color images is calculated by combining a DLT algorithm and a RANSAC algorithm; panoramic image mosaics is completed after image correction of the color images by correction geometric transition matrix; and dynamic regulation of image three-valued upper and lower thresholds and a threshold of the edge detection sobel operator is carried out during the process of panoramic image mosaics. According to the invention, image mosaic accuracy and robustness are raised.

Multi-source multi-modal activity recognition for conducting aerial video surveillance comprising detecting and tracking multiple dynamic targets from a moving platform, representing FMV target tracks and chat-messages as graphs of attributes, associating FMV tracks and chat-messages using a probabilistic graph based mapping approach; and detecting spatial-temporal activity boundaries.

The invention provides a virtual three-dimensional map establishing method and belongs to the field of electronic maps. The method comprises the steps that 1, an aerial video of a target building in atarget region is acquired, wherein the aerial video contains coordinate information, synchronously recorded, of a camera under a world coordinate system; 2, the aerial video is divided into video frames according to a time order, and two-dimensional coordinates of building image matching points in two frame images under the camera coordinate system are acquired; 3, the two-dimensional coordinatesof the building image matching points in the two selected frame images are converted according to internal parameters and external parameters of the camera in combination with the coordinate information of the camera under the world coordinate system, three-dimensional coordinates of the building image matching points under the world coordinate system are obtained, then three-dimensional reconstruction is performed on the building images, and building three-dimensional images are obtained; and 4, the building three-dimensional images are transplanted to corresponding positions in a satellitemap or a two-dimensional map or an existing three-dimensional map, and the virtual three-dimensional map containing the target building is obtained.

The invention discloses an automatic target tracking method for aerially photographed videos and aims to solve the technical problem that an existing aerially photographed video target tracking method based on stable image detection and data association is high in tracking error rate. According to the technical scheme, firstly, a pyramid optical flow method is used for extracting robust optical flow track features rather than simple motion foreground segmentation; secondly, motion constraint, position constraint and existence duration constraint are utilized for effective management of an optical flow track; lastly, clustering in time dimension is conducted according to a spatial clustering result of a plurality of continuous frames to effectively solve the problem of spatial clustering errors under the conditions of target intersection and a close range between targets. According to the automatic target tracking method for the aerially photographed videos, a target tracking result is obtained on the basis of spatial clustering, and therefore the tracking error rate is lowered. Tests prove that the tracking error rate is lowered by 10% from 18% in the background art to 8%.

The present invention discloses a self-encoding neural network-based wind turbine visual detection system. According to the system, a part of frames of wind turbine images in an unmanned aerial vehicle aerial video are pre-processed, and the pre-processed images constitute a training sample set and a test sample set for wind turbine visual detection; a self-encoding neural network is built; positive and negative samples in the training sample set are arranged in a self-encoder so as to be pre-trained; an activation quantity which is obtained through calculation is adopted to replace original input, so that a new sample feature set is formed; new feature vectors are inputted to an established BP neural network, so that the two-pass classifier of a wind turbine can be obtained; the sample images in the test set are arranged in the classifier so as to be tested; areas containing the wind turbine are searched through using a random sliding window method; and screening is performed throughnon-maximum suppression, so that a final wind turbine detection area can be obtained. The system of the invention can be widely applied to fields such as pattern recognition and intelligent inspectionof wind power generation.

The invention relates to an expressway vehicle detection method based on improved visual background extraction, which comprises the steps of 1, initializing a read-in image, wherein an input color image is converted into a grayscale image; 2, carrying out background molding on the image, carrying out background initialization if the image is a first frame image, and otherwise, carrying out background updating so as segment out the background; 3, segmenting out a road surface from the foreground; 4, carrying out eight-neighborhood filling on the image; 5, removing a lane line by using a characteristic that the lane line is narrow; 6, filling the internal part of a vehicle by using a longitudinal filing method; and 7, extracting a vehicle area and drawing an external rectangular frame of the vehicle area. According to the invention, vehicles on the expressway in aerial video of an unmanned aerial vehicle are detected by adopting a mode of image processing, the cost is low, the detection precision is high, the real-time performance is good, and the application range is wide.

The invention discloses an aerial video highway lane line detection method based on line spacing characteristic point clustering, which comprises the following steps: reading video frame; Image segmentation; equilibrium from a large number of data points, fully extract the characteristic points of each lane line; the similarity matrix is constructed based on the similarity measure method of line spacing, and the characteristic points of different lane lines are clustered. The cubic B-spline model of lane is established and the parameters of the model are estimated by the improved RANSAC algorithm; correction and prediction of lane parameters. The invention detects and pretreats the edge of the expressway in the video of the aerial shooting of the expressway, Reducing the processing time ofthe next frame, effectively removing the interference pixels outside the freeway, has a better clustering effect on lane feature points, can get more accurate and more stable lane fitting effect, andcan achieve real-time processing effect.

The invention discloses an intersection vehicle queuing length detection method based on an aerial video. The intersection vehicle queuing length detection method comprises the following steps: step 1, registering the aerial video; step 2, setting a vehicle queuing length detection region at an intersection entrance path; step 3, detecting a red light state; step 4, when a lane is at the red light state, detecting moving vehicles in the pre-set vehicle queuing length detection region by using a three-frame difference method; step 5, determining a vehicle queuing region; step 6, detecting the quantity of static vehicles of the vehicle queuing region and calculating an actual vehicle queuing length by using an actual distance Scale represented by a calibrated image unit pixel; and step 7, repeating the steps to realize continuous vehicle queuing length detection. According to the intersection vehicle queuing length detection method, the vehicle queuing length detection is carried out by the aerial video and mutual shielding and size changes of the vehicles in the monitored video are overcome, so that an intersection vehicle queuing condition is directly represented and the detection precision is extremely improved.

Provided are an aerial stereoscopic video streetscape system based on a navigation airship and an implement method. The system acquires aerial video streetscape by a viewing mode combining automatic flying and manual flying of a navigation airship, and solves a difficulty in streetscape data acquisition and update at high efficiency and low cost. A rotor wing airship captures aerial panoramic videos, and divides layers according to a flying height and by using latitude and longitude as coordinates, thereby forming the aerial video streetscape of a city. The system has wider view and higher height, displays the stereoscopic space of the city in an omnibearing way, effectively prevents a case that current ground streetscape is restricted by lack of roads, obstruction of buildings, ground fluctuation, and blind angles, may cover any area of the city in the air, replaces static streetscape with dynamic video streetscape, and may display the real space really and dynamically in a large data bulk.

The invention discloses an unmanned aerial vehicle real-time video sharing method and system. The sharing method comprises the following steps: (1) starting up aerial photographing equipment or image photographing equipment; (2) photographing by taking a media stream timestamp manner and/or a text timestamp manner and encoding photographed data to generate a media stream; (3) transmitting the media stream to a cloud server through a communication network for carrying out real-time sharing of data. With the adoption of the unmanned aerial vehicle real-time video sharing method and system, real-time sharing and processing of unmanned aerial vehicle images/videos is realized, the speed is rapid and the efficiency is high; meanwhile, all-directional and high-precision sharing of a plurality of types of information is also realized.

The invention discloses a lane line dynamic detection and lane boundary fitting method, which comprises the following steps of: performing lane line detection from an image frame of a road aerial video, marking a lane line, and fitting a cubic parabolic equation of the lane line. The method comprises three parts of extracting aerial road video frames, training a generative adversarial network, detecting lane lines and fitting a lane line equation. The method comprises the following steps: firstly, extracting an image frame and manually marking a lane line with a special color; training a deepnetwork based on conditional adversarial, and detecting lane lines in the test set; and finally, extracting coordinates of special color pixel points on the lane line, and fitting by using a least square method to obtain a cubic parabolic equation of the lane line. The method is applied to lane line detection and fitting in an aerial road video, the detection performance of the method is verified,the result shows that the method has high accuracy for lane lines at different positions, and a good detection effect is obtained.

The invention discloses a driver lane-changing behavior analysis method based on an aerial photography video vehicle trajectory. Based on the image tracking technique, the movement video data of a vehicle, collected by the aerial photography of an unmanned aerial vehicle, are extracted and processed. After that, through the data fitting, the mapping and the conversion of coordinate systems, a corresponding movement track graph, a displacement space-time diagram in the direction of lanes and a lane deviation fluctuation diagram perpendicular to the direction of lanes are obtained. According tothe invention, a vehicle trajectory processing and conversion method is provided by the invention. By the method, vehicle trajectories in a video can be efficiently, conveniently and accurately extracted and processed. The method has important significance on the research and the development of trajectory analysis-based relevant traffic phenomena and driver characteristics.