71 results about "Adaptiv" patented technology

The invention discloses a tunnel water leakage area detection and recognition method based on infrared and visible light image fusion. The method comprises the following steps: firstly, collecting infrared and visible light data in a tunnel of an area to be detected by utilizing an industrial camera; preprocessing the acquired data to obtain initial infrared and visible light image data, and registering the infrared and visible light images; fusing the infrared and visible light data obtained after registration by using a deep learning technology; carrying out target detection of water leakageand elimination of oxygen resin abnormity on the obtained fusion image to realize detection and recognition of water leakage in the tunnel; and finally, an inertial navigation system is used for achieving the function of positioning and marking the water leakage area in the tunnel where an external positioning system cannot be used. The method effectively solves the problem of interference of poor illumination conditions in the tunnel and abnormal epoxy resin on water leakage detection, has the advantages of high precision and high adaptability, and provides a basis for backtracking and rechecking of diseases in the tunnel.

The invention discloses a centroid identification and positioning method for an infrared facula image. The method comprises the following steps: taking an image of an infrared light spot centroid areaneeding to be identified and positioned as an interested area image; performing depth conversion and gray linear stretching on the region-of-interest image to generate a preprocessed image; performing adaptive threshold calculation on the preprocessed image to obtain a binarization threshold of the preprocessed image; binarizing the preprocessed image, finding out an object contour in the regionaccording to a contour tracking algorithm, and selecting a maximum contour point set as a light spot contour; and positioning the mass center of the infrared light spot according to the light spot contour. According to the centroid recognition and positioning method for the infrared light spot image, the binarization threshold obtained through the adaptive threshold method changes along with imagefeature transformation, object contour information can be better extracted, the adaptability is higher, and the anti-noise capacity is higher. And a selected contour tracking algorithm and a maximumcontour point set are selected as light spot contours, so that the light spot contours can be well determined, and the speed is higher.

The invention relates to a large-scale centralized photovoltaic power station image data acquisition method based on a multi-rotor unmanned aerial vehicle (UAV). A UAV can process visible light videostream information by carrying a processor; after the UAV is lifted off, the position and arrangement direction of photovoltaic module strings in the field of view can be calculated in combination with the own state information of the UAV, and the UAV can realize the automatic tracking of the photovoltaic module strings according to the calculation result; by calculating a flight distance, the UAVcan automatically collect visible light images and infrared image data; when the UAV reaches the regional boundary, a steering strategy is automatically executed; and by repeating the above process,the UAV can realize an image data collection task for the entire centralized photovoltaic power station, and can be further used for fault detection. Compared with other UAV-based data acquisition methods, the method of the invention is simple and efficient to implement, does not depend on the position information of photovoltaic modules, only needs to obtain regional information of the photovoltaic power station, and has strong environmental adaptability.

The invention provides multi-effect and low foam organic radar cooling liquid and application thereof, and relates to the technical field of cooling liquid. The radar cooling liquid is prepared from the following raw materials in parts by weight: 44-95 parts of ethylene glycol, 5-56 parts of deionized water, 1.0-4.5 parts of an organic composite corrosion inhibitor, 0.1-0.5 part of a pH buffer agent, 0.08-0.16 part of a rubber protective agent, 0.01-0.02 part of an antisludging agent, 0.00023-0.00028 part of a coloring agent and 0.002-0.003 part of defoaming agent. The cooling liquid generatesless foam, can achieve corrosion protection of multi-metal materials and solders of liquid cooling systems such as radars, has good adaptability with non metals, and has long service life. The cooling liquid has good application effects in liquid cooling systems such as military radars, airborne warning and control systems, aircraft carriers, electronic equipment and guided missiles, and specialair conditioners, central air conditioners, refrigeration houses, armoured vehicles, refrigeration equipment, vehicles and other low temperature environment equipment cooling systems.

The invention provides a mining locomotive pedestrian detection method based on multi-information fusion. The method comprises: establishing pedestrian data sets of visible light and infrared light inan underground coal mine; performing CLAHE finite contrast adaptive histogram equalization and optimal notch denoising method processing based on a third-order Butterworth function on the two kinds of images for the underground special environment; adjusting and improving a YOLOv3 target detection network by adopting dense connection and a multi-scale pooling structure; extracting and fusing features of the two types of images; optimizing a loss function by adopting a cross entropy function; model establishment, improving the accuracy, the real-time performance and the stability of the modelin an underground complex environment. Meanwhile, the training precision of the network is further improved by adopting a transfer learning method, and the training time is shortened, so that the pedestrian target is detected by using the target detection model obtained by training, the pedestrian detection result in front of the mining locomotive is output in real time, and the detection speed required by the running of the mining locomotive can be met.

According to the invention, the antenna polarization requirement of a multi-information-amount satellite-borne SAR imaging system is met, reasonable unit layout combination is adopted, the problem ofmutual coupling between two polarization energies of the sub-array is isolated by using the metallized through holes, the defects of poor cross polarization and poor port isolation of a satellite-borne active phased array radiation sub-array are overcome, the space environment adaptability and reliability of the radiation sub-array are fully considered, and the aspects of material selection processing, processing selection, structural design layout and the like are fully emphasized, so that the radiation sub-array has the advantage of large-scale engineering application.

A camouflage system adaptively matches the visible and IR spectrum of surrounding vegetation. A bio-chromophore dye solution circulated through an upper channel and distributed by a pulp or fabric matches the visible and biological IR spectrum, while water evaporated from the upper channel or from a separate lower channel matches the water IR spectrum. Dye can be retained in the pulp or continuously circulated. Permanently printed colors and / or patterns can also be included. Petalation cooperative with a channel flow pattern can release evaporated water and inhibit LiDAR detection. An upper waxy layer and surface embossing can avoid specular reflections. The camouflage signature can be compared with the environment and automatically adjusted as needed. Embodiments include thermal management, electromagnetic shielding, and / or radar scattering / absorbing subsystems. An airbag ground plane can match a terrain contour and avoid LiDAR detection. Multiple zones can provide adaptive display of visible and IR patterns.

The invention provides a medium wave infrared detector imaging time delay test system and method. The system comprises a portable computer, a control recorder, a target plate (a medium wave infrared LED array) and a medium wave infrared detector, wherein the portable computer is a notebook computer and is connected with the control recorder through the Ethernet cable, a control port of the targetplate and a detector signal output port are connected with the control recorder through special cables. The system is advantaged in that a detector output video signal can be timely acquired, throughhigh-precision control on an LED array work state and real-time processing on acquired images, medium wave infrared detector imaging time delay detection is realized, the detection information is further outputted to the computer to display dynamic imaging time delay change in real time, and high precision, strong environment adaptability and good readability, portability and robustness are realized.

The invention provides an infrared target detection method, which comprises the following steps: compressing an input infrared image, and roughly screening an image sliding window by using windows with different sizes; quantizing the continuous data into binary codes by using a local sensitive hash and iterative quantization method; outputting an optimal rotation matrix and a binary coding matrix;calculating a Hamming distance between the candidate window and the target template to obtain a first round of candidate window; mapping the first round of candidate window into the original infraredimage, and carrying out fine screening; wherein the window with the Hamming distance smaller than the threshold value is the position of the target. According to the technical scheme, the technical problems of low detection probability, low speed and poor complex scene adaptability in an existing infrared target detection algorithm are solved.

The invention relates to an optical light splitting device for receiving three wave bands through a common window and application of the optical light splitting device in handheld, vehicle-mounted andairborne distance measurement camera shooting white light observation equipment and distance measurement double-path camera shooting equipment. The device comprises a common lens group, a composite prism, a beam splitter prism, an avalanche transistor detector, a near-infrared detector and a visible light detector. External light passes through the common lens group and is refracted into two paths of light by the composite prism, wherein one path of light enters the avalanche transistor detector, and the other path of light enters the beam splitter prism and then is refracted into two paths of light, and the two paths of light respectively enter the near-infrared detector and the visible light detector for imaging. The device is simple in structure, small in size and light in weight, achieves the three functions of distance measurement, visible light imaging and near-infrared light imaging at the same time, and is wide in application range and high in adaptability.

The invention relates to a method and system for eliminating unsupervised vehicle re-identification deviation based on synthetic data. The method comprises the following steps: S1, converting the synthetic data into a pseudo target sample having a similar style with a target domain through a consistent generative adversarial network SPGAN, and carrying out the pre-training of a model through the pseudo target sample; s2, performing an unsupervised domain adaptive or completely unsupervised task through a pre-trained model; s3, calculating the image direction and camera similarity through a pre-trained direction model and camera model, and further obtaining the final vehicle similarity for testing. According to the method and the system, the vehicle re-identification performance is improved, and the adaptability is high.

The invention discloses a monocular self-adjusting fire point three-dimensional positioning method and device, which solves the problems that the binocular position is fixed, the solving precision islimited, the shooting position cannot be adapted when being shielded, and the equipment cost is high in the existing binocular image method. According to the method, a thermal infrared imager is usedas an image acquisition means, and a ball screw is driven by a stepping motor to drive monocular image acquisition equipment; a controller is used as a control means, a control system is compiled at aPC end, and a corresponding camera calibration calculation algorithm, a corresponding shooting distance calculation algorithm and a corresponding thermal image fire point three-dimensional coordinatecalculation algorithm are matched, so that automatic early warning of the fire point is realized, and meanwhile, three-dimensional space coordinates of the fire point are given. The method and the device can be applied to three-dimensional positioning of fire points in a limited space such as a room or a vehicle or a ship, are strong in environmental adaptability and positioning accuracy, and caneffectively save the cost.

The invention provides an infrared long-wave cut-off filter and a preparation method thereof, and a film system structure of the infrared long-wave cut-off filter is MLMHM|Sub|(MH)<^>11M, Sub is a substrate, (MH)<^>11M is a short-wave pass filter film system, MLMHM is an antireflection film system, H is a Ge film layer, M is a ZnS film layer, and L is a YbF3 film layer. The infrared long-wave filter is high in transmittance, accurate in cut-off position, good in chemical stability and high in environmental adaptability, the cut-off depth is smaller than 0.01, the infrared long-wave cut-off optical filter can be widely applied to an infrared detection optical system, the infrared long-wave cut-off optical filter manufactured through the method is cut off in the 9.6-11 micron waveband, and the average transmittance T of the 7.5-9.0 micron waveband is larger than or equal to 95%. The environmental adaptability of the infrared long-wave cut-off filter meets the national military standard of optical films.

The invention discloses a self-obstacle-avoiding and fixed-point-tracking spherical aircraft which mainly comprises a spherical framework shell, a gyroscopic stabilizing system outer ring, a gyroscopic stabilizing system inner ring, a motor supporting frame A, a paddle A, paddle tacking knobs, a brushless motor A, a socket cap screw A, a pan-tilt seat, a motor supporting frame B, a brushless motor B, a paddle B, a control system single chip microcomputer, an infrared sensor, a platform, a brushless motor C, a paddle C, a motor supporting frame C, a brushless motor D, a paddle D, a motor supporting frame D, a socket cap screw B, a socket cap screw C, a battery, tee couplings, carbon fiber tubes, a pin A and a pin B. The self-obstacle-avoiding and fixed-point-tracking spherical aircraft runs flexibly without a dead point, is light in weight, small in size and simple in mechanical structure, has better flexibility and adaptability, can be controlled remotely, has an automatic obstacle avoiding function, is provided with an expansion interface in a reservation manner, and can expand a miniature camera and a WIFI (wireless fidelity) module later.

The invention discloses a multi-feature fusion complex background mid-infrared weak and small target detection method and system, belongs to the technical field of infrared target detection and recognition, and solves the problems that in the prior art, when a complex background mid-infrared weak and small target is detected, the algorithm adaptability is poor, a large number of false alarms occur in a detection result, and the precision of the detection result is affected; the method comprises the following steps: firstly, extracting radiation characteristics, multi-order directional derivative characteristics and spectrum characteristics which respectively represent radiation characteristics, structural characteristics and intensity characteristics of a weak and small target, fusing the multiple characteristics, constructing a characteristic saliency map, and inhibiting background noise while enhancing the target; using a CFAR adaptive detection method to calculate a segmentation threshold value of the image, obtaining a binary segmentation result, carrying out morphological processing to screen out false targets caused by isolated points and noise, and obtaining a final weak and small target detection result; the method is low in algorithm complexity, high in adaptability to complex backgrounds, high in detection precision and convenient for engineering implementation.

The invention discloses a method for converting a thermal infrared image into a visible-light-like color image based on a CycleGAN, and belongs to the field of image processing based on a generative adversarial network. The implementation method comprises the following steps of: according to image domain conversion information of a thermal infrared image and a visible light image in a database, connecting two generators for converting the thermal infrared image into a color visible light image and converting the color visible light image into the thermal infrared image by utilizing the cyclic consistency loss of a cyclic generative adversarial network CycleGAN; high-precision pairing of the thermal infrared image and the color visible light image is not needed, and the purpose of unsupervised learning from the thermal infrared image to the color-like visible light image is achieved. And a full convolutional neural network structure is further adopted in the network of the generator, so that the size of the input thermal infrared image is not limited. According to the learning of the image domain conversion information of the thermal infrared image and the visible light image in the database, the natural color rendering of the thermal infrared image is realized, and the observation adaptability and the target identification and detection speed are improved.

The invention discloses a lens for infrared temperature measurement. The lens comprises an outer shell and an outer glass cylinder; a wireless transceiving module is fixed on the inner side surface ofthe outer shell; an inner attaching plate is fixed in the outer shell; an electric push rod is fixed to the rear side of the inner attaching plate. The output end of the electric push rod penetratesthrough the inner attaching plate and is fixedly provided with an inner lens body; an inner glass cylinder movably penetrates through the front side of the outer shell, the front end of the inner lensbody movably penetrates through the inner glass cylinder, a driven gear is arranged at the front end of the inner glass cylinder in a rotating mode, an outer attaching plate is fixed to the side faceof the driven gear, and a first motor is fixed to the left side of the outer attaching plate. According to the lens for infrared temperature measurement, the lens can be protected, the protective cover used for protecting the lens can be conveniently replaced, the temperature measurement angle of the lens can be enlarged through reflecting of infrared rays, then multi-directional temperature measurement can be conducted on the external environment, the lens can be conveniently installed on a workbench with a plane, a round face and the like, and adaptability is high.

The invention discloses an SPCP infrared weak and small target detection method based on 3DATV constraints and belongs to the field of infrared image processing and target detection. The method comprises the following steps of: step 1, constructing the third-order tensor of an original image; 2, constructing the priori weight tensor of the original image; 3, constructing a target function by usinga tensor kernel function and a tensor l1 norm and combining with the 3DATV constraints, converting an original target detection problem into an SPCP problem, and solving the target function by usingADMM to obtain a background tensor and a target tensor; 4, reconstructing a background image and a target image according to the background tensor and the target tensor; and 5, segmenting the target image and outputting a target detection result. With the method adopted, the defect that an existing method is low in noise robustness, weak in edge inhibition force, poor in complex scene adaptabilityand the like are eliminated; the maintaining capacity of the shapes of targets is enhanced, and the accuracy of target detection is improved.

The invention relates to the field of outdoor fixing of infrared detectors, in particular to an assembly facilitating outdoor fixing and supporting of an infrared detector, which comprises a main body, a storage box, a supporting mechanism, a self-locking mechanism, a supporting rod mechanism, a telescopic mechanism and a steering mechanism, through the arrangement of the supporting mechanism, a supporting effect on the detector is conveniently achieved, the detector is prevented from shaking in severe windy weather, the adaptability of the detector is improved, through the arrangement of theself-locking mechanism, opening and closing of the cover plate and the storage box are facilitated, the detector is sealed in the storage box, and the detector is prevented from making contact with dust and other sundries in a non-working state; through the arrangement of the supporting rod mechanism, the cover plate can be conveniently adjusted and fixed to a proper angle, the controllability ofthe detector is improved, through the arrangement of the telescopic mechanism, the height of the detector is conveniently and freely adjusted, the detector is prevented from being blocked by a shelteron the ground, and through the arrangement of the steering mechanism, the detector body can be conveniently and finely adjusted to a proper direction; objects at different spatial positions can be conveniently observed by the detector main body.

The invention discloses a three-rotor-wing unmanned aerial vehicle attitude detecting system and a detecting method based on PIR (passive infrared receiving). The system comprises a PIR sensor array,a PIR signal filtering and amplifying circuit and a PIR attitude information processing center, wherein the PIR sensor array comprises a PIR detecting sensor and a PIR standard sensor. The detecting method comprises the steps of front-end data collection, middle data filter conversion, data operation and attitude angle solution. Compared with the prior art, the system disclosed by the invention can serve as an attitude detecting system which can be independently utilized, so that a three-rotor-wine unmanned aerial vehicle has stronger adaptability to various flying environments and especiallycan finish tasks of attitude detection, flying control and the like in the absence of the support of an attitude detecting system like a gyroscope and the like. No matter being utilized as a main detecting method of an unmanned aerial vehicle or applied to additional calibration detection design, the attitude detecting system disclosed by the invention can well improve reliability and stability ofunmanned attitude detection.

The invention discloses a multifunctional window with a windproof lighting function. The multifunctional window comprises two rod set rotating devices, a scissor type telescopic device, a windproof rainproof device and a controller; during operation, the multifunctional window is connected with the controller through a network, a Bluetooth or infrared remote control, and the controller controls the rod set rotating devices to rotate to a specified angle, so that the scissor type telescopic device and the windproof rainproof device are driven to rotate, and the scissor type telescopic mechanismis controlled to be elongated or shortened to a designated position; when in severe weather such as typhoon rainstorm, the rotatable scissor type telescopic device enables the multifunctional windowto be attached to the outer wall and elongated, so that windproof and rainproof functions can be achieved; in the general case, the control rod set rotating devices and the scissor type telescopic device can also be controlled to stay in the proper position, so that the effect of adjusting lighting is achieved; and the multifunctional window is suitable for various environments, can be intelligently controlled, and is high in reliability, good in adaptability and convenient to operate.

The invention provides a low-conductivity dielectric cooling liquid for an ethylene glycol type phased array radar, and an application of the low-conductivity dielectric cooling liquid, and relates tothe technical field of cooling liquids. The radar dielectric cooling liquid is prepared from the following raw materials in parts by weight: 66 to 68 parts of ethylene glycol; 32 to 34 parts of deionized water; 2.0 to 4.5 parts of an organic nonionic composite corrosion inhibitor; 0.08 to 0.16 part of an organic non-ionic rubber protective agent; 0.01 to 0.02 part of a scale inhibitor; and 0.002to 0.003 part of a defoaming agent. The cooling liquid is low in conductivity, capable of achieving corrosion protection of multi-metal materials and welding flux of liquid cooling systems such as radars, good in non-metal adaptability and long in service life; and the cooling liquid has a good application effect in liquid cooling systems such as military radars, airborne radars, early warning aircrafts, aircraft carriers, electronic equipment, guided missiles and the like.

The invention discloses an infrared weak and small target detection method based on integrated fusion features, which relates to the field of machine learning, and comprises the following steps: acquiring an initial image of an infrared weak and small target as a training set, and establishing a classifier to obtain a trained model; obtaining a to-be-detected image and filtering the to-be-detected image through a high-pass filter; performing constant false alarm threshold segmentation on the filtered image; carrying out candidate target area marking on the segmented binary image, and calculating the center coordinate of the binary image; according to the center coordinate of each candidate target, image blocks are taken from the to-be-detected image; extracting feature parameters of each to-be-detected image block; and classifying the feature parameters of the to-be-detected image block through the trained model, obtaining and outputting the center coordinate of the target, and completing target detection. According to the method, the classification capability of the fusion features is higher, the classification precision can be improved, the convergence speed can be increased, and thus the purpose of reducing classifier parameters is achieved; the method has enough adaptability to complex application scenes, and is convenient for engineering application.

The invention provides a wide-area multi-target radar mounting device. The wide-area multi-target radar mounting device comprises a main mounting seat, wherein positioning holes are formed in the two sides of the surface of the main mounting seat; and an extension seat, wherein one side of the extension seat is fixedly mounted on one side of the main mounting seat through a fixing bolt. The wide-area multi-target radar mounting device provided by the invention is provided with the elevation angle adjusting bolt which can adjust the elevation angle of the multi-target radar body in the elevation angle adjusting hole and can be finely adjusted under a certain acting force, the multi-target radar is convenient to mount, the adaptability to the mounting environment is high, the performance of the multi-target radar can be conveniently exerted to the maximum extent, the wide-area multi-target radar mounting device is simple in structure and suitable for different types of traffic road sections, the multi-target radar can be installed on the cross rod road section or the vertical rod road section, the protective baffle reduces the vibration amplitude of the multi-target radar body after being impacted by hail through the buffer structure at the bottom, and therefore the adaptability of the multi-target radar to the installation and use environment is improved.

The invention provides an infrared camera weak light environment compensation method and device and electronic device. The infrared camera weak light environment compensation method is based on a target detection and recognition system, and comprises the steps that a camera performs image collection; the target detection and recognition system obtains a target area from an image; the target detection and recognition system detects the brightness of the target area; the target detection and recognition system judges whether a difference value between the brightness value of the target area and an ideal brightness value meets a preset value or not; if yes, the target detection and recognition system recognizes an image in the target area; if not, after the brightness of the infrared light supplement lamp is adjusted, image collection is carried out again through the camera. According to the weak light environment compensation method for the infrared camera, the brightness of infrared imaging of the target in the weak light environment can be improved, the imaging quality problem caused by software and hardware differences of various cameras is avoided, the adaptability of a video image processing algorithm is improved, and application scenes are enriched.

The invention provides an infrared weak and small target real-time detection method and system based on a target detection model, and the method comprises the steps: constructing an infrared weak and small aircraft detection data set, and carrying out the redesign of a feature extraction backbone network, so as to enable the network to be more suitable for the feature extraction of a weak and small target; a spatial pyramid pooling layer is added behind the backbone network again to obtain different receptive fields, so that the positioning capability of the network model to weak and small targets is enhanced; and finally, aiming at the relatively fixed and single size of the infrared weak and small target, reducing detection heads, and fusing shallow layer information and deep layer information through a characteristic pyramid tower, so that the network has higher detection accuracy and stronger robustness. The weak and small target detection accuracy and speed under the complex background are improved, and the robustness and adaptability of a weak and small target detection algorithm are improved.

The invention provides a remote sensing alarm system and device for a garbage power plant, and the system comprises an infrared module, a sensing module, an alarm module and a control module, and forms a tight data flow direction in the system. The infrared device, the electronic fence device, the alarm device and the garbage grab bucket apply the system; the infrared device, the electronic fence device, the alarm device and the garbage grab bucket are in data interconnection through a wireless network; the infrared device is located on the discharging platform, and the electronic fence device and the garbage grab bucket are located in the garbage pool. When the transport vehicle is detected, the alarm device gives an adaptive alarm according to different position states of the garbage grab bucket and the electronic fence device; the transport vehicle runs into the sensing area of the infrared device in the unloading process, and the alarm device gives an alarm to a worker in the garbage pool, so that the operator adaptively controls the garbage grab bucket, the problems of collision and covering of the garbage grab bucket in the prior art are avoided, and the working safety is improved.

The invention discloses an infrared perspective imaging detecting chip which a ceramic shell, a metal support, a heat dissipation board, a drive control and perspective image preprocessing module, an area array non-refrigeration infrared detector and an area array infrared refraction micro lens. The drive control and perspective image preprocessing module, the area array non-refrigeration infrared detector and the area array infrared refraction micro lens are coaxially and sequentially arranged in the ceramic shell. The rear portion of the ceramic shell is arranged on the top of the metal support and the top of the heat dissipation board. The drive control and perspective image preprocessing module is arranged at the joint of the rear portion of the ceramic shell, the metal support and the heat dissipation board. The area array non-refrigeration infrared detector is arranged on the top of the drive control and perspective image preprocessing module, and the area array infrared refraction micro lens is arranged on the top of the area array non-refrigeration infrared detector. The infrared perspective imaging detecting chip is compact in structure and has the advantages that single-chip perspective imaging detection is performed on an infrared target based on an infrared beam direction, the measuring precision is high, the infrared perspective imaging detecting chip is easily compatible or coupled with a conventional infrared optical system, and the adaptability of the target and the environment is good.