39results about How to "Narrow down the area" patented technology

The invention relates to a pedestrian detection method based on foreground analysis and pattern recognition, relating to the technical field of image processing. The method comprises the following steps: adopting a Gaussian mixed model to carry out background modeling on a video image and using threshold operation and morphology after-treatment to extract the foreground of the video image; using a contour characteristic and a pedestrian height prior model to analyze the foreground and obtain a preliminary pedestrian detection result; sampling nearby the position of the preliminary pedestrian detection result, using a pedestrian pattern recognition classifier to further judge a sampling area and removing an inaccurate preliminary pedestrian detection result so as to obtain a final pedestrian detection result. The pedestrian detection method not only can improve the degree of accuracy of pedestrian detection, but also can increase the processing speed of pedestrian detection in a video and can be applied to dynamically variable complex occasions.

The present invention discloses a person searching method, apparatus and system. The method comprises: receiving an image that contains a to-be-searched person; performing segmentation processing on the image, so as to obtain a person whole-body region, a person upper-body region and / or a person lower-body region; separately extracting various types of feature information of images in the person whole-body region., the person upper-body region and / or the person lower-body region; separately calculating a similarity degree between each type of feature information of each region and corresponding feature information in a database; performing a weighted operation according to a preset weight value of the similarity degree, so as to obtain a fusion similarity degree of images in the person whole-body region, the person upper-body region and / or the lower-body region that correspond to each person image in the database; and separately obtaining searching results of person images, which is sorted according to the fusion similarity degree, in the database. The method, apparatus and system disclosed by the present invention have the advantages of an accurate retrieval result and high retrieval efficiency.

The present application provides an image processing method and device. The method comprises the following steps that: at least one target vehicle image is determined, wherein the target vehicle images contain license plates, the true colors of the license plates contain white; white reference pixels are obtained from the license plate areas on the at least one target vehicle image; white balancegain is acquired according to the white reference pixels; and white balance correction is performed on the at least one target vehicle image through using the white balance gain. With the method adopted, white reference points in images can be determined accurately and efficiently; color adjustment is performed on the images based on the white reference points; and therefore, the colors of the images are more accurate.

The invention discloses a node localization method based on centroid iterative estimation, and belongs to the technical field of wireless sensor networks. The method comprises the following steps: firstly, determining an iterative convergence condition of the method according to a position relationship between an unknown node and a region enclosed by initial communication beacon nodes; calculating a centroid coordinate of a region enclosed by current communication beacon nodes and a distance between the centroid coordinate and the unknown node through coordinates of the communication beacon nodes and received signal strength indications (RSSIs) between the coordinates of the communication beacon nodes and the unknown node; and replacing the communication beacon nodes being farthest from the unknown node by calculated centroid nodes in order to reduce a region in which the unknown node is located. The aims of reducing the region in which the unknown node is located and increasing the node localization accuracy are fulfilled through multiple iterations.

The invention provides an interconnection structure and a forming method of the interconnection structure. The interconnection structure comprises a semiconductor substrate, a metal layer on the semiconductor substrate, an interlayer medium layer on the metal layer, and a discrete metal wiring layer formed inside the interlayer medium layer; the metal wiring layer is a top metal layer with an interconnection structure; the metal wiring layer is divided into a brim area and an interior area, and the brim area and the interior area are isolated through the interlayer medium layer. According to the technical scheme, the area range of electric connection between the metal wiring layer and below adjacent metal layer is reduced by connecting the brim area of the metal wiring layer and below adjacent metal layer, so that the degeneracy phenomenon of a gate oxidation layer due to the fact that plasma generated when a metal welding plate layer in a semiconductor interconnection structure is fabricated penetrates into the substrate can be reduced.

The invention discloses a wireless sensor network positioning method and a wireless sensor network positioning system. The method comprises the following steps that an accumulation result of strength values of signals collected by a sensor is received, the positions of a plurality of objects are respectively assumed to be the center of a grid where the signals are located, signal strength of the plurality of objects is calculated, a reconstruction matrix is built, signals sent from the plurality of objects are reconstructed by solving the smallest and optimization problems of l 1 norm to obtain a reconstruction result X, a first reconstruction error is obtained through calculation and compared with a preset value, and when the first reconstruction error is smaller than and equal to the preset value, the positions of the plurality of objects are confirmed. According to the method, the sensor just needs to collect a small amount of data, hardware requirements of the sensor are low, network communication quantity is reduced, and the service life of the network is prolonged. In addition, the method can be used for positioning of the plurality of objects, and can further reduce the area range of the plurality of objects using an iteration backtracking algorithm, and positioning precision is improved.

The invention discloses an image processing method and device and electronic equipment. The method comprises the steps of detecting multiple key points corresponding to a processing object in an image, and according to the key points, determining a to-be-processed area corresponding to the processing object; judging whether all area boundaries of the to-be-processed area go beyond an object boundary of the processing object or not; if yes, determining the area boundaries going beyond the object boundary of the processing object as target boundaries; adjusting the target boundaries to shrink the area range of the to-be-processed area, and determining the shrunk to-be-processed area as an actual processing area; aiming at each pixel point in the image, judging whether the pixel point belongsto the actual processing area or not, if yes, processing the pixel point according to a preset object processing rule. As can be known, by means of the image processing method, it can be guaranteed that the boundaries of the actual processing area do not go beyond the boundary of the processing object, and the pixel points which need to be processed are divided through the actual processing area.In this way, the calculation amount can be reduced, and the processing speed is increased.

The invention discloses a polygonal cutting blade with bulge parts in a hole and a cutter. The polygonal cutting blade comprises a blade body consisting of an upper surface, a lower surface and a plurality of side surfaces connected with the upper surface and the lower surface, wherein a center hole penetrating through the upper surface and the lower surface is formed in the blade body, the bladebody is in central symmetry about the central axis of the center hole, each side surface and the upper surface intersect to form a plurality of cutting edges, one bulge part is arranged in the position, corresponding to each side surface, of the inner surface of the center hole, and the bulge parts directly face the corresponding side surfaces. The cutting blade has the characteristics of positioning reliability and stability, multiple clamping and long service life. The cutter comprises the cutting blade, fastening screws and the center hole of the cutting blade produce over-positioning multidirectional clamping under multiple constraints, deformation directions and areas unfavorable for fasteners are effectively controlled, the vibration of the cutting blade can be effectively attenuated during cutting, and the service life of the cutter is prolonged.

The invention belongs to the technical field of radio signal positioning, and particularly relates to a multi-target direct positioning method based on neural network computing. The multi-target direct positioning method includes the steps: firstly, decomposing an interested target area into a plurality of sectors, and detecting an existed target sector by the aid of a multilayer feed-forward neural network; secondly, further decomposing the sector into a plurality of sub sectors when a plurality of targets in some sector is detected to ensure at most one target in each sub sector, and detecting the sub sectors with the targets by the aid of the multilayer feed-forward neural network again; thirdly, sequentially performing spatial filtering on the sectors or sub sectors with the targets bythe aid of the multilayer feed-forward neural network; finally, independently, parallelly and directly positioning the targets in different sectors or sub sectors by the aid of a radial basis neuralnetwork.

The invention puts forward a discharge port dust control device. The discharge port dust control device comprises a tubular discharge port, a dust control baffle, a dust collection pipeline and air suction equipment, wherein the dust control baffle is annularly arranged at a certain distance above the discharge end of the tubular discharge port, a dust control opening is arranged between the dust control baffle and the tubular discharge port, one end of the dust collection pipeline is connected with the dust control opening from the upper portion of the dust control baffle, the air suction equipment is connected with the other end of the dust collection pipeline, and air suction is conducted through the dust collection pipeline so that negative pressure can be formed below the dust control baffle. The discharge port dust control device is simple in structure and low in cost, and restrains dust from rising while guaranteeing quick discharge.

The invention discloses a circular blade with an accurate positioning structure and a cutting tool. The circular blade comprises a blade body, wherein the blade body is composed of an upper surface, a lower surface and a side face which is used for connecting the upper surface with the lower surface; and the blade body is provided with a center hole which penetrates through the upper surface and the lower surface, the blade body is centrosymmetric relative to the central axis of the center hole, the upper surface intersects with the side face to form a cutting edge, the inner surface of the center hole is provided with at least two lug bosses, and the included angle between every two adjacent lug bosses is smaller than 180 degrees. The cutting blade has the advantages that positioning is reliable and stable, multi-ply clamping can be achieved, and the service life is long. The cutting tool is provided with the circular blade, sliding or rotating of the circular blade in the peripheral direction in the using process can be eliminated, the stability of the cutting edge in a cutting region is guaranteed, other cutting processes are prevented from participating in one cutting process, the usable frequency of the circular blade on the cutting tool is effectively stabilized, and then the service life of the cutting tool is prolonged.

The invention discloses a loopback detection method, related equipment and device, and the method comprises the steps: obtaining a current key frame of point cloud data and the position information of the current key frame; determining a sub-region corresponding to the current key frame based on the position information of the current key frame, the sub-region being obtained by dividing the target region; responding to a complete historical key chain existing in the sub-region corresponding to the current key frame, and matching the complete historical key chain with the current key frame; wherein the complete historical key chain is composed of at least one continuous historical key frame; and determining that loopback detection is successful in response to successful matching of the current key frame and the complete historical key chain. By means of the mode, the loopback detection accuracy can be guaranteed while the real-time performance of loopback detection is guaranteed.

The lithium battery fireproof heat insulation structure comprises a protection device, the protection device comprises a U-shaped protection outer frame, side protection buffer mechanisms, a top protection buffer mechanism and a linkage buffer mechanism, the side protection buffer mechanisms are installed on the inner circumferential side face of the U-shaped protection outer frame in a sliding mode, and the top protection buffer mechanism is installed on the inner top of the U-shaped protection outer frame; the linkage buffer mechanisms are in sliding fit with the side protection buffer mechanisms and the top protection buffer mechanisms on the adjacent surfaces respectively, so that the lithium battery can achieve fireproof and heat insulation effects, impact force generated when the lithium battery explodes in a fire can be relieved, the range of the explosion influence area of the lithium battery is reduced, and the safety of the lithium battery is improved. And through the early warning system of the lithium battery fireproof heat insulation structure, the duration of the lithium battery spreading from the current stability maintaining state to the thermal runaway state is estimated, the reliability prediction of the duration of the thermal runaway of the lithium battery is realized, the actual duration of the lithium battery from the stability maintaining state to the thermal runaway state is furthest approached, and the reliability of the lithium battery is improved. And reliable early warning prompt before the lithium battery is out of control is achieved.

The invention discloses a crowdsourcing fingerprint database construction method based on map information screening and matching, comprising the following steps: speculating pedestrian motion information by using an inertial sensor in a smart phone, uniquely determining a trajectory starting point through two-layer screening in combination with the limitation of map obstacles, and matching the position of each step of a path through particle filtering. And in combination with RSS data collected in the walking process, an offline fingerprint library is rapidly and accurately constructed, and online positioning is assisted. According to the method, the fingerprint database can be constructed more quickly and densely, and the positioning precision is higher.

The invention relates to the field of environment-friendly planting equipment, in particular to an environment-friendly plant planting mechanism for a building wall surface. The mechanism comprises a device body, a planting cavity is formed in the device body, a rotating ring is movably installed on the front face of the device body, a driving plate is movably installed at the outer side of the rotating ring, and a moving frame is movably installed at the upper end of the device body. A trimming blade is fixedly installed on the front face of the rotating ring, the rotating ring can drive the trimming blade to cut off the part, exceeding the exterior of a plant shape when rotating, an automatic trimming effect is achieved, the trimming blade is bent in an arc shape, and cutting edges are arranged at the two sides of the trimming blade, so that the greening plant modeling space in a planting cavity can be reserved; and meanwhile, no matter what direction the airflow drives the rotating ring to rotate, the trimming blade can cut off the plants, airflow grooves are formed at the two sides of the driving plate and are arranged in a continuous obtuse-angle bending manner, so that the airflow can move along the airflow grooves after making contact with the driving plate, and then the acting force of the airflow on the driving plate is improved.

The invention discloses a severe cold region courtyard layout optimization design method. The method comprises the following steps of 1, establishing a benchmark geometric model and a design scheme comparison group of severe cold region village and town courtyard layout; 2, setting boundary conditions for village and town courtyard microclimate simulation in severe cold areas; step 3, implementingsimulation analysis of village and town courtyard microclimate in severe cold regions, and constructing an index scoring grid diagram; 4, courtyard behavior paths and areas of village and town residents in severe cold areas are collected, and a behavior track grid diagram is drawn; 5, superposing the microclimate index scoring grid graph and the town resident behavior track grid graph to form anevaluation area superposed graph; and 6, evaluating the microclimate index performance of the overlapped area. The design method provided by the invention is more targeted and implementable, the efficiency of solving the microclimate environment problem of the village and town courtyards in severe cold regions can be improved, and more possibilities are provided for courtyard layout design.

The embodiment of the invention provides a road traffic sign recognition method and apparatus. The road traffic sign recognition method includes the steps: acquiring the shot images; utilizing line segments with inclined characteristics to search triangular signs or circular signs from the shot images; and according to the found triangular signs or circular signs, recognizing the road traffic signs of the warning type or the forbidden type. The road traffic sign recognition method and apparatus can avoid the process of searching the signs or images with horizontal or vertical characteristic from the shot images, thus improving the searching speed, reducing the area range for recognition, reducing the recognition objects, improving the recognition speed, and accelerating recognition of the road traffic signs as a whole, and can more timely prompt users according to the recognized road traffic signs, thus enabling vehicle driving to be more safe.

The invention provides a touch display device. The touch display device comprises a touch panel, a display panel and a glue sealing compound structure, the glue sealing structure is arranged between the touch panel and the display panel and defines a glue injection area, and the glue sealing structure comprises a plurality of first retaining walls and a plurality of second retaining walls. The first retaining walls are arranged on the periphery of the glue injection area and at least define a first edge, a second edge, a third edge and a fourth edge, the first edge and the second edge are opposite to each other, the third edge and the fourth edge are opposite to each other, and the third edge and the fourth edge are arranged to form at least one notch respectively. The plurality of first retaining walls are formed by double faced adhesive tape, and the plurality of second retaining walls are arranged on the inner sides of the plurality of first retaining walls and define at least one diversion trench with the plurality of first retaining walls.

The invention relates to a pedestrian detection method based on foreground analysis and pattern recognition, relating to the technical field of image processing. The method comprises the following steps: adopting a Gaussian mixed model to carry out background modeling on a video image and using threshold operation and morphology after-treatment to extract the foreground of the video image; using a contour characteristic and a pedestrian height prior model to analyze the foreground and obtain a preliminary pedestrian detection result; sampling nearby the position of the preliminary pedestrian detection result, using a pedestrian pattern recognition classifier to further judge a sampling area and removing an inaccurate preliminary pedestrian detection result so as to obtain a final pedestrian detection result. The pedestrian detection method not only can improve the degree of accuracy of pedestrian detection, but also can increase the processing speed of pedestrian detection in a video and can be applied to dynamically variable complex occasions.

The invention discloses a TOF camera-based passenger flow counting method and a device. The method comprises the following steps of 1, acquiring a depth image; 2, screening a detection region througha height threshold value to obtain a mask image; 3, performing morphological treatment on the mask image; 4, acquiring a detection candidate region of a passenger head in a color image according to the mask image; 5, positioning the passenger head in the candidate region by adopting a depth neural network model; 6, tracking by KCF through combining Kalman filtering; 7, counting the passenger flow.According to the invention, not only the problem that the pattern recognition function cannot be realized by adopting a depth image is solved, but also invalid regions in the image are filtered through the depth image. Therefore, the detection speed and the detection accuracy are accelerated. In conclusion, the purpose of quickly and accurately counting the passenger flow can be achieved.

The invention provides a ghost point detection method and device and a storage medium, and the method comprises the steps: determining a target region according to the position of a currently detected contact after the contact is detected on a screen; detecting whether the contact jitters in the target region; and if the contact jitters in the target region, determining that the contact is a ghost point. According to the method and the device, the accuracy of ghost point detection is improved, so that adverse effects of ghost points on terminal operation can be effectively avoided, and the availability is high.

The invention relates to a medicine bottle cleaning device in medical laboratory, which comprises a cleaning table and a telescopic mechanism arranged above the cleaning table. A cleaning device is provided on the side of the telescopic mechanism close to the cleaning table; A hollow rod arranged on the telescopic mechanism, the other end of the hollow rod is movable in the axial direction, and a trapezoidal rod with one end extending to the inner side of the hollow rod is provided; the present invention improves the traditional way of cleaning and fixing medicine bottles , the method of using the fixed part and the cleaning part separately is converted into a cleaning and fixing integrated type. During the cleaning process, the expansion and lowering of the telescopic mechanism is used to drive the contact between the cleaning device and the medicine bottle. The pressing down of the inner bottom wall of the bottle and the pressing down of the pressing part on the top of the bottle realizes the double fixing effect on the medicine bottle, realizes the fixing of the medicine bottle, and then carries out the subsequent cleaning steps to realize the fixed area occupied by the fixing parts. Reduction also reduces its production cost.

The invention relates to a light emitting structure and a clock with the same. A circuit device and light emitting sources are accommodated in a frame of the light emitting structure; a plurality of light guide holes corresponding to positions of the light emitting sources are formed in a light guide casing covering the frame so as to help the light emitting source to irradiate light to an outer part of the frame via the light guide holes, and light spots can be projected on a wall surface for shadow arrangement of a household environment or time indicating of the clock.