101 results about "Aerial photo" patented technology

The invention relates to a system and method for generating a lane-level navigation map of an unmanned vehicle based on multi-source data. The lane-level navigation map comprises an offline global map part and an online local map part. According to an offline module, within a target region where the unmanned vehicle runs, original road data is acquired through satellite photos (or aerial photos), a vehicle sensor (laser radar and a camera) and a high-precision integrated positioning system (a global positioning system and an inertial navigation system), then the original road data is subjected to offline processing, multiple kinds of road information are extracted, and finally the road information extracting results are fused to generate the offline global map. The offline global map is stored through a layered structure. According to an online module, when the unmanned vehicle automatically drives in the target region, the road data in the offline global map is extracted according to real-time positioning information, and the online local map with the vehicle as the center within the fixed distance range is drawn. The system and method can be applied to fusion sensing, high-precision positioning and intelligent decisions of the unmanned vehicle.

The invention discloses an optimal route selection method for overhead transmission line route. Onboard lidar equipment is adopted to acquire laser point cloud data and aerial digital photo data of the transmission line route corridor range; the onboard lidar data is processed after the wave filtering and the sorting of the laser point cloud data, and the points of the ground surface are made into a digital elevation model with high precision; then data processing is carried out by utilizing the data of the digital elevation model with high precision and ortho-rectification is carried out to the aerial photo by utilizing the internal and external orientation elements of the aerial digital photo to generate digital orthophoto maps; through the overlying of the digital elevation model and the digital orthophoto maps, the tridimensional visualization of the transmission line route corridor can be realized to optimize the transmission line route selection; finally prearrangement of power pole and power pole arrangement are carried out according to the data of plane cross sections. The route selection platform of the invention has simple operation and lifelike tridimensional scene, thus being convenient for full roaming and multi-view observation and greatly improving the efficiency of the inner plane cross section survey operation. Compared with the optimal route selection technology based on the aerial photographing measuring method, the efficiency of the inner plane cross section survey operation can be improved by about 75 percent.

The invention discloses a local Retinex dehazing algorithm based on HSV color spaces. The local Retinex dehazing algorithm is used for conducting image enhancement on an image with fog being unevenly distributed under the foggy day condition. The method includes the following steps that A, the multi-layered foggy day image is processed in different color spaces, so that the image contrast ratio is enhanced, and the image definition is improved; B, the size of the image is obtained, the sizes of subblock images are judged in a self-adaptation mode, and a luminance image of the foggy day image is partitioned; enhancement processing is conducted on all sub images of the luminance component, and the image definition is improved; C, the area proportion factors of all pixel points in the overlapped sub images are calculated, weighted summation is conducted on all the overlapped sub images, and image fusion is achieved; D, the saturability component of the image is processed, and the image color is restored. By means of the local Retinex dehazing algorithm, scenery details at different levels can be restored clearly and quickly, the visual effect of the image is improved, and the algorithm is suitable for aerial photo systems at different regions.

The invention discloses a tree three-dimensional reconstruction method based on unmanned aerial vehicle aerial photo sequence images. According to the tree three-dimensional reconstruction method, firstly, characteristic regions of a crown are extracted in a watershed partition method, then, the characteristic regions are matched by calculating region correlation coefficients in an RGB color space, matched characteristic dot pairs are extracted, depth information is calculated through the binocular stereo vision principle combined with imaging models of an aerial photo camera, and finally a three-dimensional model of a tree is constructed based on L system laws. Compared with the prior art, the tree three-dimensional reconstruction method has the advantages that grain, color and outline information of the crown are considered, and the characteristic spot set which can show the crown complicated structure is fully extracted; in addition, the depth information of the characteristic spot set is calculated according to complicated imaging models of the aerial photo camera with shake, transverse moving and the like of the camera in the aerial photo process of an unmanned aerial vehicle considered, a reasonable tree modeling approach is adopted, and therefore reasonable approximate three-dimensional models of the tree are constructed.

The invention discloses an image matching method suitable for automatically detecting the flight quality of the aerial photography. The method comprises the steps that firstly, images are preprocessed, denoising and filtering are respectively carried out on two digital images to be matched, and the two digital images to be matched are obtained with the aerial photography measuring method in the measured area and are two-dimensional images; secondly, SIFT characteristic extracting is carried out, the SIFT characteristic extraction is respectively carried out on the two digital images through a data processor, all characteristic points of the two digital images and SIFT characteristic descriptors of all the characteristic points are obtained; thirdly, a kd tree is established to carry out the bi-directional search to determine matching points; fourthly, gross error removing is carried out according to the RANSAC algorithm. According to the method, the steps are simple, design is reasonable, the achieving process is convenient, robustness and the using effect are good, and the method can effectively solve the problem that according to an existing method, matching is prone to failure matching or error matching.

The invention provides a method of analyzing vehicle dynamic position in traffic accident; the method comprises the steps of S1, measuring, on a ground, highway dimensions and ground marking lengths by means of on-site aerial photography in order to perform integrated data acquisition; S2, acquiring monitoring video data to form a highway grid coordinate system, subjecting the grid coordinate system to calibration processing and error evaluation, and subjecting a monitoring video image containing multiple accident vehicles to grid measurement in order to determine accurate positions of the accident vehicles in the single monitoring video image; S3, analyzing travelling trajectory, speed change, braking safety distance, travelling lane and travelling direction change of each of the multiple accident vehicles, drawing the accident vehicles positioned by the grid coordinate system to a positive overlook aerial photo according to a time point sequence, and analyzing to obtain traffic accident integrated information. By drawing grid coordinates for aerial photos taken in later stage and site images about conditions of accident sites, it is possible to depict a motion trajectory of a vehicle in the traffic accident accurately.

A system featuring a server communicating with the network with game hosting software programmed on the server, a terrain database and a time sequence GPS position linked message database each stored on the server and accessed by the game hosting software is disclosed. The terrain database stores a plurality of three dimensional renderings of golf courses created from digital aerial photographs of each golf course, wherein each photo includes elevational and slope data for each aerial photo. The time sequence GPS position linked message database contains a plurality of messages collected by a portable computing device and transmitted to the server by the device during an actually played round of golf. Each message is: transmitted by the device to the server; associated with an elapsed time from the beginning of the play of the round of golf, and linked to a GPS determined position from where the message was transmitted.

The present invention provides a method of combining various types of data and software in order to arrive at a composite graphical representation of a construction site, including surface and subsurface features. A 2-D or 3-D graph of subsurface contoured surfaces is created. A 3-D wire-frame model of surface and subsurface features is created. Aerial photographs of the site can be incorporated and the aerial photographs contoured surfaces and wire-frame model are to produce a complete Visual Reduction Modeling Language (VRML) model of both surface and subsurface features. The VRML model allows the user to identify and visualize a relationship between sub-surface features and above ground features.

The invention discloses a remote sensing target detection method of a double attention mechanism, and belongs to the field of image processing in computer vision. The method comprises the following steps: 1, inputting a multi-scale remote sensing aerial image; 2, segmenting the input picture into 512 * 512 small pictures; 3, inputting the small picture into a network, and generating a feature mapthrough a receptive field module; 4, constructing a double attention mechanism convolutional neural network as a matching function, optimizing network parameters, and redefining a feature map generated through the convolutional neural network; 5, realizing classification and bounding box regression of the targets through subsequent operation of the network; and 6, obtaining a regression box with the highest score by inhibiting the non-maximum values of the plurality of regression coordinate boxes, wherein the box is an interested target position in the remote sensing aerial image. According tothe invention, the target in the remote sensing aerial photo can be effectively detected, and the accuracy and generalization of the target detection network are improved.

The present invention belongs to the technical field of geological disaster survey, and discloses an electric transmission line corridor geological disaster survey method and system. The method comprises the steps of: constructing a three-dimensional scene model graph of an electric transmission line corridor through aerial photo and location of an unmanned aerial vehicle, and collecting meteorological information and electric transmission line operation information in real time; performing classification of the three-dimensional scene model graph of the electric transmission line corridor based on a supervised classification method to identify a terrain type and a disaster generation region; taking electric transmission line poles and towers as main research subjects to perform distinguishing of change states of the poles and towers; and finally, taking the collected meteorological information, the electric transmission line operation information, suspected disaster region informationand the change information of the poles and the towers as data input to compare with a current geological disaster model library so as to perform early warning for regions where disasters are likelyto happen. The electric transmission line corridor geological disaster survey method and system can achieve object-level monitoring of the poles and the towers so as to improve the whole geological disaster survey quality and efficiency and guarantee the accurate assessment of the geological disasters.

The invention provides a road three-dimensional model modeling method based on a lane-level high-precision map. The method comprises the steps of S10, extracting road base drawings of the high-precision map according to a remote sensing image and an aerial photo, and classifying the road base drawings to serve as graphic layers; S20, extracting elevation data information at important road nodes according to a local digital elevation model; S30, importing the road base drawings of the high-precision map into Sketchup, and making a basic framework of a road three-dimensional model according to the elevation data information at the important road nodes; S40, establishing a buffer area for road signs in the high-precision map; S50, importing the established buffer area into the Sketchup, and performing three-dimensional modeling on the road signs; and S60, adding other attribute information to the built three-dimensional model. According to the method, the remote sensing image, actual aerially photographed road information and terrain information reflected by the digital elevation model are effectively combined, so that real road information can be reflected; and the three-dimensionalmodel conforms to visual habits of a driver, so that driving navigation can be effectively realized.

The invention discloses a method for detecting a moving object on the basis of online dynamic background modeling in a video. The method comprises the following steps: collecting an image and carrying out pre-processing; judging whether the image frame is the first frame, if yes, initializing a background model, or else, jumping to the next step; generating a background image by utilizing a dynamic background model; extracting matching point pairs in the current frame of image and the background image; screening the accurate matching point pair and calculating a correction parameter by utilizing RANSAC; correcting the dynamic background model to the coordinate system of the current frame of image; dividing the pixels of the current frame of image into a moving pixel and a background pixel by utilizing the corrected background model; dynamically updating the background model by using the background pixel; and carrying out morphological processing, connected component analysis and area constraint on the moving pixel. According to the method, an advanced background modeling-based moving object detection algorithm under a fixed platform can be conveniently applied to an aerial photo platform, so as to detect the moving object more accurately.

The invention discloses a montanic photovoltaic power station layout platform and layout method. The method comprises the steps that an unmanned aerial vehicle takes off according to a montanic path plan, performs photographing on mountain topography, acquires geographical location information of a mountainous region, and stores the geographical location information and aerial photos; the geographical location information and the aerial photos are imported in a background processing module, the background processing module performs recovery processing on the geographical location information and the aerial photos, and a topographic contour map is generated; and the topographic contour map is imported in three-dimensional photovoltaic software PDP, the three-dimensional photovoltaic software PDP automatically identifies contour lines and elevation point information in the topographic contour map, a three-dimensional photovoltaic power station geographical model is generated, shadow shielding information is calculated automatically according to the three-dimensional photovoltaic power station geographical model, a layout region is analyzed, the layout region is subjected to photovoltaic array layout design, support foundation design and electric connection design, and the three-dimensional photovoltaic power station geographical model is automatically converted into a montanic photovoltaic power station simulation design model. The layout platform and the layout method are high in efficiency, high in precision and low in cost.

The invention discloses a shooting device for an aerial photo holder. The shooting device is characterized by comprising two rotators, a connecting seat and a driving mechanism, the connecting seat is used for connecting the rotators, the driving mechanism is arranged on the connecting seat and used for driving the rotators to rotate on a horizontal plane, the rotators are arranged below the connecting seat, each rotator comprises a camera shell with a cavity and a camera obliquely arranged in the camera shell, a shooting hole corresponding to a lens of the camera is formed in each camera shell, the optical axis of each camera is downward, an included angle is formed between each optical axis and a vertical direction, and the driving mechanism comprises a power device and a transmission mechanism connected with the power device and the camera shells. Shooting angles can be flexibly adjusted, vibration absorption performance and sealing performance are fine, the working environment and the shooting quality of the cameras are ensured, and the shooting device can be widely applied to the aerial photo holder.

The invention discloses a moving video track and video playback synchronization displaying method in an electronic map. When a video content is photographed, GPS moving track information including coordinate and elevation data and the posture data of an IMU (inertial measurement unit) which is rigidly and fixedly connected with an aerial photo camera at a certain sampling frequency are acquired. Solving GPS data and IMU (inertial measurement unit) data are jointed to determine the field position coordinates of an image center. The index relationship between a position sequence and a video frame sequence is established, and the index relationship is used to determine the mapping between an edited video and the position information to acquire a position marker sequence. According to the position marker sequence, the marker video of a video scene moving track is recorded and displayed, and the marker video is embedded into a playback video for synchronization playback. When the video content is played, the moving video track is displayed on the electronic map.

The invention discloses a high-precision three-dimensional city modeling method, and relates to the field of modeling. The method is characterized in that firstly high-precision three-dimensional laser point cloud and high-resolution digital aerial photos of a measurement area are acquired by using airborne LIDAR, and high-precision digital elevation models and digital orthoimages are rapidly extracted and high-precision modeling of buildings is performed so that a three-dimensional city electronic sand table and three-dimensional building object frame model scenes are manufactured; secondly, acquisition of the highly dense laser point cloud and the high-resolution digital photos of the city scenes on the both sides of the measurement area is completed by using vehicle-mounted mobile laser scanning measurement, vehicle-mounted data quality optimization is performed by reference to the airborne LIDAR result and then three-dimensional modeling is rapidly performed on various types of sketch elements of the city; and finally the three-dimensional digital model results acquired by airborne LIDAR and vehicle-mounted mobile laser scanning measurement are integrated so that the quality and the working efficiency of three-dimensional city modeling can be greatly enhanced, and the project development cost can be greatly reduced.

The invention discloses a multi-camera set cooperative acquisition holder and a control system thereof. Themulti-camera set cooperative acquisition holder comprises camera sets, a control panel, a storage battery, a photographing trigger control key and a high-precision GPS module. The control panel comprises an MCU, camera set control interfaces, a power supply conversion module and a TF card. The wiring end of the MCU is electrically connected with the wiring ends of the camera setcontrol interface, the power supply conversion module and the TF card through wires. The number of the camera sets and the number of the camera set control interfaces are both N. The invention relates to the field of geographic information aerial survey technology. By means of thecooperative acquisition holderand the control system thereof, thecamera set composed of an orthographic camera and a plurality of oblique cameras can be realized. A high-precision GPS module is carried on the holder, so that the control system of the holder can automatically identify a planned operation area and calculate the flight speed of an airplane. The working number of camera sets is controlled and the shooting time interval of the camera sets is adjusted to meet the requirement for the heading overlapping degree of aerial photos.

The invention relates to the technical field of artificial intelligence detection. The invention relates to the field of aerial photography, in particular to an aerial photography building detection and segmentation method and device based on Mask R-CNN. The method comprises the steps that firstly, aerial images of urban buildings are acquired, the contours of building objects in the aerial imagesare marked, a training set and test set data are established, and a training data set is enhanced in a non-random covering data enhancement mode; constructing an aerial photo building detection and segmentation network; training the network by using the training data set, and performing test and performance evaluation on the trained segmentation model through test set data to obtain a final aerial photography graph building segmentation model; and applying the obtained model to the building aerial image needing to be processed by the user to obtain a final building aerial image segmentation image. According to the method, a deep learning method is used, the speed and the efficiency are improved, and the segmentation accuracy and the robustness of the model are improved by applying a transfer learning and non-random masking data enhancement method.

The invention discloses a method for rapidly and accurately determining the shape and area of a forest gap, which comprises the following steps: (1) determining coordinates of a forest zone to be determined; (2) controlling, by a ground processing system, according to the coordinates, an unmanned plane remote sensing system to fly to a position 200 m above the ground of the forest zone to be determined, allowing the unmanned plane remote sensing system to hover; (3) performing aerial photographing of the forest zone to be determined by the unmanned plane remote sensing system, wherein the unmanned plane remote sensing system comprises a control module (210) for receiving instructions from the ground processing system, and controlling a power module, a global positioning system module, and a shooting module based on the received data, and a shooting module for receiving instructions from the control module and performing aerial photographing based on the received instructions; (4) processing the aerial photographs, that is, processing the aerial photographs by an ERDAS software; (5) calculating the shape and area of the forest gap. The method has the advantages of rapidness, accuracy, sensitive reaction, simple structure, labor saving and time saving, and the like, and is suitable for popularization and application.

The invention relates to a transmission equipment recognition platform located on an unmanned aerial vehicle. The transmission equipment recognition platform comprises an aerial photo camera, an equipment recognition system, a wireless transceiver and a Freescale IMX6 processor. The aerial photo camera is used for shooting transformation equipment on the ground to obtain a transmission equipment image, the equipment recognition system is used for processing the transmission equipment image to recognize the varieties of transmission equipment in the transmission equipment image, and the Freescale IMX6 processor is connected with the equipment recognition system and the wireless transceiver and used for wirelessly sending the varieties of the transmission equipment to a far end power supply unit management platform through the wireless transceiver. A haze removing processing device is further used in the transmission equipment recognition platform to overcome the influence of haze weather to the image. By means of the transmission equipment recognition platform, the varieties of the transmission equipment can be accurately and quickly recognized in the sky in real time even under the haze weather.

The present invention provides an aerial photo image processing method, an unmanned plane, and a head-mounted display device and system. The aerial photo image processing method comprises: obtaining the visual field change coordinates of a head-mounted display device, wherein the visual field change coordinates indicate the user visual field changing of the head-mounted display device; and capturing the image of the field angle corresponding to the visual field change coordinates in the image shot by the unmanned plane, and sending the image of the field angle to the head-mounted display device. According to the invention, the problem can be eliminated that a user is vertiginous caused by transmission time delay of the image.

The invention discloses a method for managing an engineering site based on unmanned vehicle aerial photography and a building information model, and the method comprises the steps: firstly building athree-dimensional building information model of the engineering site, and building engineering construction stage progress models of different stages according to an engineering progress plan and an actual completion condition; a timing unmanned aerial vehicle aerial photography system is established, and the standardization and effectiveness of aerial photography are guaranteed; performing three-dimensional display on the progress condition of the building information model; and inserting the aerial photo into an engineering construction stage progress model, adjusting the position size to enable the image to be aligned with the model, finishing an engineering site construction management model for engineering site management, and implementing monitoring management of an engineering construction site according to the model. According to the method, the aerial photography of the unmanned aerial vehicle and the building information model are combined, the same-platform browsing of the aerial photography of the unmanned aerial vehicle and the building information model is realized, the correspondence between the aerial photography of the unmanned aerial vehicle and the spatial position of the building information model is ensured, and the real-time monitoring of an engineering site construction process is realized.

A method of producing aerial photograph data is provided which is capable of positively utilizing stored photograph data and an image of a desired object included in a photographed image. The method includes a step of determining a distance between a position of a virtual photographic principal point (virtual photographic principal point position) obtained by the use of spatial position information of a target point and spatial direction of a view vector and a position of an actual photographic principal point (actual photographic principal point position) of each of a plurality of aerial photographs to automatically select a specific aerial photograph, and a step of extracting, from the specific aerial photograph, a selected image having the target point arranged at a predetermined position therein.