47results about How to "Reduce search area" patented technology

The invention provides a rapid distance-measuring method for pedestrian on a road ahead on the basis of an on-board binocular camera and discloses a binocular detecting and distance-measuring method for a pedestrian applicable to an automatic driving system. The method comprises the steps: training a pedestrian detecting model by means of a pedestrian detecting method based on study, and detecting pedestrians in a binocular scene by means of the pedestrian detecting model; calculating a matching point of a point, chosen from the detection result of a left image, in a right image according to a color characteristic and a scale-invariant characteristic, and calculating a parallax value between matching pairs; obtaining an accurate parallax value of the detection result by means of a parallax mid-value searching method, and calculating a distance between the pedestrian and the camera according to a geometrical relationship between a camera coordinate system and an image coordinate system. With respect to the problems of slow calculation speed and inaccurate distance measurement in the present pedestrian detecting and distance-measuring method for a pedestrian based on the binocular camera, the method accelerates the detection speed by means of a road column-like model; and the distance-measuring accuracy of pedestrians in a road scene is improved by means of the matching of many characteristic points.

The invention relates to a saliency detection and LSD linear detection-based airport extraction method comprising the following steps: an original airport detection data set is built; remote sensing images are extracted one by one from the data set and are subjected to downsampling operation, Frequency-tuned method based saliency detection is conducted on the downsampled images, a full resolution saliency map is formed, linear segments in the saliency map are obtained via LSD linear segment detection operators, the obtained linear segments are segmented and then connected to form long linear segments, parallel straight lines are screened out, obtained parallel straight lines are clustered, a minimum bounding rectangle zone for each type is extracted and then expanded outward by 1.1 times to form a final airport zone, and coordinates of the airport zone are output and drawn up in original remote sensing images. Via LSD linear detection in the airport extraction method disclosed in the invention, texture and edge information in the images can be successfully extracted, FT saliency detection is combined with the LSD linear detection, detection is expedited, and precision is ensured while detection accuracy is enhanced.

The present invention discloses a rapid fingerprint identification method based on the singular point topology structure, and the rapid fingerprint identification method can fast process and identify the fingerprint images in the automated fingerprint identification system by using singular point topology structure in fingerprints. The present invention classifies the fingerprint database according to the classification property and position stability of the singular point in fingerprint images, and selects the most reliable singular point as a reference point to build the singular point topology structure of fingerprints and a module to identify; the present invention firstly detects the authenticity of singular points in a fingerprint image, and then classifies the database according to the relative position of singular points to build a topology structure adjacent to the singular point; and the topology structure can be used to build the fingerprint module to speed up the comparison process of the fingerprints; the rapid matching of fingerprints in the automated fingerprint identification system is a difficulty of the fingerprint recognition technology; the rapid fingerprint identification method, on one hand, can guarantee the recognition performance, on the other hand, can significantly improve the recognition speed of the automated fingerprint identification system, and can widen the scope of application; moreover, the method reforms the prior system slightly.

The invention discloses a dynamic environment-friendly travel path planning method for automobiles. The method comprises the following steps that an actual vehicle-mounted experiment is completed, anemission calculation model is established, a Dijkstra algorithm is optimized, an ArcGIS vector map is formed, path planning simulation is carried out, specifically, the actual vehicle-mounted experiment is completed by using MAHA METDH6.3 so as to acquire real-time vehicle emission data, a passenger vehicle emission calculation model is established based on VSP by adopting the data, the road network vector map is built according to a real urban road map by utilizing an ArcGIS platform, road network data of the vector map is imported into MATLAB, a dynamic optimal path is simulated and solved by combining the optimized Dijkstra algorithm with the emission calculation model and dynamic traffic information simulation, the established dynamic environment-friendly travel path planning model canplan the route schemes of different optimization targets better, the difference of different routes and different indexes is judged, the method is applied to actual road driving, drivers are assistedto complete the optimal path planning, and environment-friendly travel path planning suitable for being coupled with dynamic traffic is realized.

The invention discloses a satellite communication in motion rapid capturing method; the method comprises the following steps of 1, performing system initializing; 2, establishing a coordinate system;3, carrying out pre-adjustment of antenna pointing; 4, carrying out coarse capture of an antenna; and 5, carrying out accurate alignment of the antenna. The method is simple in steps, reasonable in design and convenient to implement, and pre-adjustment of antenna pointing is realized by utilizing the position and speed information of a carrier detected by a GPS module, so that the target communication satellite can be rapidly captured under different motion states of the carrier, the accurate alignment of the antenna and the target communication satellite is realized, the capture and fine alignment time is short, the stability of the antenna relative to the inertial space is ensured, the rapid response capability of the communication in motion system is improved, and the method is high inpracticability.

An embodiment of the invention provides an information transmission method of an unmanned device. The information transmission method relates to the technical field of aircrafts, and can determine geographic position information of a target terminal through acquiring a terminal signal in a coverage range by arranging a base station on the unmanned device. The method comprises the steps that: the base station obtains signals of terminals, and finds out the signal of the target terminal; the base station determines the geographic position information of the target terminal according to the signal of the target terminal; and the base station transmits the geographic position information to a flight control device, so that the flight control device performs flight control over the unmanned device according to the geographic position information. The information transmission method of the unmanned device and the base station provided by the invention can be applied to the flight process of searching or performing other tasks by the unmanned device.

The invention discloses a determination method of variable granularity of integer ambiguity of a single-frequency receiver in the absolute positioning, belonging to the technical field of satellite navigation and positioning and relating to a method for carrying out the single-point precise positioning by the single-frequency receiver in a satellite navigation positioning system with the utilization of the satellite carrier phase. The determination method is based on the utilization of the single-point static positioning LAMBDA algorithm to calculate the float solution and the distribution variance of the integer ambiguity, firstly carries out the rough integral mapping of the float solution to complete the roughening of the integral domain of search space, then utilizes the rough integral granularity to change the search region and carries out the search of the integer ambiguity according to search conditions to obtain the integral domain of the integer ambiguity; then the rough integral granularity is continuously reduced, the process is repeated till the rough integral granularity is equal to 1, then the search of the integer ambiguity is completed, and the precise carrier phase integer ambiguity is obtained. The use of the invention can rapidly, conveniently and reliably realize the carrier phase high-precision positioning of the single-frequency receiver and can be applied in the low-cost and high-precision navigation positioning system.

The invention discloses an object prediction area optimization method applicable to target identification, comprising the following steps: expanding an object prediction area; performing image segmentation; performing super pixel bounding box calculation and super pixel significance evaluation; and finally, getting an optimized object prediction area based on sliding window search of super pixels. By controlling the size of super pixels, the scale of image segmentation is increased, the search scope of super pixels is reduced, and therefore, the time complexity of the algorithm is low and is only related to the number of super pixels in an image; as the pixels in the super pixels are consistent and segmentation of the local edge is better, a better locating effect can be produced by taking super pixels as the basic elements of sliding window search compared with single pixels; the search area is reduced effectively, and the traditional pixel-based sliding window search algorithm is accelerated; and in addition, by cascading the method, the target identification accuracy of the existing target identification algorithm can be improved.

The invention belongs to the technical field of embedded computer image processing and in particular relates to a real-time runway detection method based on sensor prior knowledge. A conventional runway detection algorithm mainly aims at satellite images and downward-looking high-altitude aerial images, the images are not accordant with fore-sight aerial images required by a vehicle-mounted synthetic vision system, in addition, detection is carried out for static images generally, a relatively long calculation time is needed, and the real-time requirement cannot be met. Therefore, the invention discloses the real-time runway detection method based on sensor prior knowledge. By adopting the method, information such as flying positions and postures provided by a vehicle-mounted sensor is also included into calculation, so that search areas of runway detection can be greatly reduced, and positions of runways in images can be rapidly and accurately detected.

The method discloses a crowd statistical method based on depth learning. The method extracts a motion foreground from a video, ensures camera angle of view and perspective invariance by using a humanbody region model, and finally determines the human body region statistical number through preprocessing, extraction, and detection. The method can not only reduce the search area of the sliding window and improve the search efficiency, but also overcome the deformation defect of the monitoring video, caused by view angle, monitoring scene distance, and the like, and installation and deployment ofthe system are simple, a detection model based on a depth learning convolution neural network improves the accuracy of human detection, the use of a non-maximum suppression method eliminates redundant sub-regions to reduce repeated counting, so that the results of human detection, human location and population statistics are more accurate.

Provided is a population global optimization method for regional dynamic subdivision based on local Lipschitz estimation. The population global optimization method is combined with the Lipschitz estimation theory under the framework of a population evolutionary algorithm and comprises the steps of constructing a Lipschitz estimation lower bound bearing surface by extracting neighborhood information of new individuals to achieve dynamic subdivision of a search region; effectively identifying part of an invalid region by means of lower bound estimation information and guiding population update; performing local reinforcement by means of the general declining direction of a linear fit convex envelope, and performing secondary subdivision on the search region according to evolution information; reinitializing part of poor individuals in a population according to lower bound estimation selection information. The population global optimization method for regional dynamic subdivision based on local Lipschitz estimation is low in computing cost, and high in rate of convergence and reliability.

The invention relates to a photovoltaic multi-peak power point tracking method based on a territory particle swarm. The method comprises the following steps: 1) adjusting a duty ratio of a PWM as 0 toobtain output voltage of a Boost circuit; 2) setting the number of the particle swarm; 3) computing an adaptive value of each particle, re-judging and selecting a global optimal particle and the global optimal particle territory according to the obtained adaptive values; 4) determining a maximum value estimation range in the particle territory; 5) judging whether the condition that the maximum value estimation range contains the adaptive value of the global optimal particle is tenable; 6) judging whether the condition that the maximum distance dmax between the particles is less than 0.02Uocaris tenable; 7) computing each particle adaptive value, re-judging and selecting the global optimal particle; 8) judging whether the condition that the maximum distance dmax between the particles is less than 0.01Uocar is tenable; and 9) maintaining the output voltage as the voltage of the obtained global optimal particle location, and judging whether the environment changes suddenly.

A system for reducing a search area for identifying correspondences identifies an overlap region within a first match frame captured by a match camera. The overlap region includes one or more points of the first match frame that are associated with one or more same portions of an environment as one or more corresponding points of a first reference frame captured by a reference camera. The system obtains a second reference frame captured by the reference camera and a second match frame captured by the match camera. The system identifies a reference camera transformation matrix, and/or a match camera transformation matrix. The system defines a search area within the second match frame based on the overlap region and the reference camera transformation matrix and/or the match camera transformation matrix.

The invention discloses an isoline iteration gravity matching algorithm based on mahalanobis distance. The real-time error of gravity measurement is considered, the navigation error covariance matrix is utilized to represent the precision of the gravity anomaly background image and the gravity real-time measurement precision, the Mahalanobis distance is adopted as an iteration objective function, the covariance matrix is introduced to represent the density degree of the point sets, the same point set has high similarity, more correct matching point pairs are reserved, the matching precision of the algorithm is improved, and the navigation and positioning precision of the system is further improved. Meanwhile, a strip search area range improvement strategy is provided based on a measurement error and an objective function value, so that the search area is reduced along with the increase of the number of iterations, the convergence of the ICCP algorithm is embodied, the calculated amount of the algorithm when the nearest point on the isoline is searched is gradually reduced, and the possibility of searching the global optimal point is increased.

The invention discloses a visual tracking method for a mobile robot. According to the method, a full-connection twin network tracking algorithm based on HOG feature similarity calculation and Kalman filtering feedback is used, HOG features based on color statistics are used as an appearance model, a plurality of similar targets in a scene are further distinguished, and identity jump of the targetsis reduced. Meanwhile, a Kalman filtering motion model is adopted, so that accurate prediction of the motion trail of the target is realized, the boundary frame is refined, the search area is reduced, the target shielding problem in the target tracking process is solved, and the calculated amount is reduced. And finally, the bounding box position information of the tracking target is transmittedto a visual servo system for controlling the robot. According to the method, the waste of computing resources caused by searching useless areas is effectively reduced.