26 results about "Camouflage" patented technology

A camouflage defeat and testing system (1200) includes a network platform (1202) and one or more edge devices (1204). The platform (1202) and / or the edge device (1204) receive inputs from one or more sensor devices (1210-1213) monitoring an area of interest (1214). For example, the area of interest 1214 may be a battlefield or staging area of an adversary. In this case, the area of interest (1214) includes a target (1216) utilizing a camouflage technology (1218) such as multispectral camouflage. The system (1200) employs spatial and / or temporal processing to process images and the sensor devices (1210-1213) to defeat the camouflage technology (1218). The system (1200) can also be employed to test and optimize multispectral camouflage capabilities.

The invention discloses a hyperspectral camouflage fiber fabric based on a flocking technology. The hyperspectral camouflage fiber fabric comprises an ultra-black fluff layer, an adhesive layer, a near-infrared absorption layer and an ultra-black base cloth layer which are sequentially arranged from top to bottom. Through a multi-layer synergistic structure, the fitting degree of the material and an ocean background spectrum at the wave band of 400-2500nm reaches 0.91 or above, the characteristics of weak reflection of visible light and strong absorption of near infrared light are reproduced, the problem of poor matching precision is solved, hyperspectral reconnaissance is avoided, and the concealment of ocean equipment is improved. The invention has the advantages of strong environmental adaptability, easy mass production, humidity and heat resistance of grade 1, no abnormity at-20 DEG C to 50 DEG C, and color fastness to sunlight of grade 5, and adapts to complex marine environment; based on a mature flocking technology, complex equipment is not needed, large-scale production can be achieved, and stability and economical efficiency are both considered.

The invention belongs to the field of video detection, and particularly relates to a method and system for evaluating the dynamic camouflage effect of a target. According to the scheme, the method comprises the following steps: acquiring information, cleaning data, loading an obtained video clip, and marking a real mask of a target in a video frame by frame; loading the obtained video clip, skipping to a first frame where the target to be analyzed appears, marking the target to be analyzed, and marking a start frame number and a stop frame number of the target to be analyzed; carrying out reading, feature extraction and feature processing on the whole target clip video, outputting result features and converting the result features into a prediction mask of a target; and calling a visual detection index evaluation tool, loading a real mask of the target, processing high-dimensional data information, and calculating to obtain a camouflage quantitative score. The method and the device provided by the invention can effectively identify the dynamic camouflage effect of the target, and have practical application values in the fields of computer vision, military confrontation technology development, camouflage animal ecological protection, camouflage animal behavioral research and the like.

The invention provides a camouflage animal detection method based on frequency domain decoupling and edge information fusion, relates to the technical field of image processing, and uses an image frequency domain decoupler to dynamically separate low-frequency (global structure) / high-frequency (edge details) features through a learnable Gaussian filter bank so as to obtain a camouflage animal detection result. And the method is efficiently used in an edge contour predictor and a camouflage target detector, so that the detection precision of the camouflage animal is effectively improved. In a high-frequency edge fuzzy scene such as a dead leaf butterfly scene, the target contour response is enhanced by a high-frequency path, and the omission ratio is remarkably reduced; and a low-frequency path reserves the overall form of the target, and the recognition robustness of large-scale camouflage targets such as zebra groups is improved. Meanwhile, under the interference environment of rain and fog / night and the like, the influence of the noise of the sensor on the frequency domain characteristics is small, and the mode of replacing multi-modal data through frequency domain decoupling enables the detection performance to have high detection precision and stability.

The invention belongs to the technical field of information security, and discloses a camouflage disturbance method and system for a large correction model, and the method comprises the steps: receiving triple input of a question, a standard answer and a student answer, and generating a standardized representation matched with the correction model; generating camouflage confrontation input by adopting lexical-level disturbance and semantic-level disturbance on the basis of the standardized representation; and performing effect evaluation on the camouflage confrontation input, and outputting a disturbance result based on a composite measurement index. According to the method, a complete process from triple input to adversarial disturbance generation and effect evaluation is realized.

The application provides a camouflage animal detection method based on frequency domain decoupling and edge information fusion, and relates to the technical field of image processing. The application uses an image frequency domain decoupler to dynamically separate low-frequency (global structure) and high-frequency (edge details) features through a learnable Gaussian filter group, and efficiently utilizes the features in an edge contour predictor and a camouflage target detector, thereby effectively improving the detection accuracy of camouflage animals. In high-frequency edge blur scenes such as dry leaf butterflies, the high-frequency channel enhances the target contour response, and the missing detection rate is significantly reduced. The low-frequency channel retains the overall shape of the target, and improves the recognition robustness of large-scale camouflage targets such as zebra groups. At the same time, in interference environments such as rain, fog and night, the frequency domain features are less affected by sensor noise. By replacing multi-modal data with frequency domain decoupling, the detection performance has high detection accuracy and stability.

The application belongs to the technical field of information security, and discloses a camouflage disturbance method and system for a large-scale correction model, which comprises the following steps: receiving a question, a standard answer and a student answer three-tuple input, and generating a standardized representation suitable for a correction model; based on the standardized representation, generating a camouflage adversarial input by adopting word-level disturbance and semantic-level generation; evaluating the effect of the camouflage adversarial input, and outputting a disturbance result based on a composite measurement index. The application realizes a complete process from input three-tuple to adversarial generation and effect evaluation.

The invention discloses a camouflage object recognition probability determination method and device and a nuclear power facility area camouflage object evaluation method and system, and relates to the technical field of camouflage. The camouflage object identification probability method comprises the following steps: acquiring a to-be-processed image; extracting a first feature set of the first target region from the to-be-processed image, and extracting m second feature sets in one-to-one correspondence with the m first background regions; determining feature similarity and vectors according to the first feature set and the second feature set; determining a first feature weight vector according to the feature similarity and the vector; the first feature weight vector is used for representing the weight of each feature in the first feature set; and determining the recognition probability of the camouflage object according to the feature similarity sum vector and the first feature weight vector. According to the embodiment of the invention, the method can improve the accuracy and efficiency of determining the recognition probability of the camouflage object.

The invention relates to the technical field of photography, in particular to photography equipment and a photography method.The photography equipment comprises a pre-embedded barrel, a disguise barrel is fixedly connected to the upper end of the pre-embedded barrel, a plurality of observation openings are formed in the disguise barrel, a circular through groove is formed in the upper end of the pre-embedded barrel, two sliding rods are fixedly connected into the pre-embedded barrel, and a supporting plate is slidably connected between the two sliding rods; the two sliding rods are sleeved with springs, the springs are located between the embedded cylinder and the supporting plate, the lower end of the supporting plate is fixedly connected with a driving motor, an output shaft of the driving motor is fixedly connected with a camera, the upper end of the camera is rotationally connected with a fixing block, the embedded cylinder is fixedly connected with a plurality of extension plates, and each extension plate is provided with a mounting hole. And a positioning pin is inserted into each mounting hole, so that an automatic protection function is achieved when wild animals are shot.

The application relates to the technical field of target detection, and discloses a multi-stage refined camouflage target detection method based on wavelet enhancement recognition, which makes up for the defects that the existing method cannot effectively utilize global information, leading to missing of detection details, and cannot process multi-layer information, resulting in feature redundancy problems, so that the method can better complete a position target detection task. First, a visible light image is preprocessed through a pre-training model, a pyramid operation is performed to expand the feature quantity, the amount of calculation is controlled while the amount of information is ensured, part of the initial features are selected for feature enhancement, and the features are screened and identified through a wavelet identification enhancement module; the highest layer features are selected and sent into an asymmetric convolution mamba module to obtain features with global information; then, a feature reverse decoder composed of multiple guide units is used to process the global information features and the identification features to obtain a prediction map, and the model is trained; when a specific detection task is performed, the trained model is used to obtain the prediction map, and the detection task is completed.

The application discloses a green bionic camouflage material simulating optical and infrared performance of vegetation and a preparation method of the green bionic camouflage material. The green bionic camouflage material is made by double-sidedly dipping and coating green camouflage paint on a microporous film. The green camouflage paint is made of the following raw materials in percentage by weight: titanium white 5-6%, near-infrared high-reflection black 9-10%, organic yellow 17-18.2%, water-absorbing resin 1.2-1.5%, composite adsorbent 1-1.2%, film-forming material 35-40% and solvent 20-35%. The composite adsorbent is made of porous white carbon black and inorganic hydrated salt. The green bionic camouflage material can simulate the spectrum of green vegetation in 380nm-2500nm and the water absorption peaks near 1450nm and 1930nm. The water-absorbing resin can lock water to simulate the inherent water of vegetation leaves; the composite adsorbent is driven by environmental temperature and humidity and solar radiation, and can realize water vapor adsorption / desorption through the microporous film to simulate the transpiration of green vegetation leaves. The diurnal radiation temperature difference between the green bionic camouflage material and vegetation leaves is within ±2℃, and the green bionic camouflage material can well simulate the infrared radiation characteristics of green vegetation.

The invention provides a camouflage object detection method based on a deep learning uncertain form. The method comprises the following steps: acquiring the spectral wavelength and spectral energy of flames when hydrogen and chlorine are combusted in different proportions through a spectrograph; performing data preprocessing on the spectral wavelength and the spectral energy; inputting the preprocessed spectral wavelength and spectral energy into a constructed deep learning network model for training; in the training process, the training round and the learning rate are adjusted until an optimal weight file is obtained; and performing real-time prediction on spectral wavelength and spectral energy output by the spectrometer interface through the optimal weight file and the deep learning network, and regulating and controlling the ratio of chlorine to hydrogen according to a prediction result. The method provided by the invention effectively overcomes the defects of a manual observation method and a distributed control system, and further improves the preparation efficiency of hydrogen chloride.

The invention discloses a dynamic camouflage type wild animal channel monitoring device and method based on multi-modal perception in the technical field of wild animal research. The device comprises monitoring equipment and a channel shielding assembly arranged in an animal channel. The monitoring device comprises a dynamic simulation piece and a multi-mode monitoring assembly arranged on the dynamic simulation piece. The channel shielding assembly comprises a storage assembly, a flexible blocking piece arranged in the storage assembly and a driving piece used for driving the storage assembly to release the flexible blocking piece. The multi-modal monitoring assembly is installed in the dynamic simulation part, the dynamic simulation part is used for hiding the multi-modal monitoring assembly, interference of the monitoring device to animals is reduced, the driving part changes the length of the portion, located on the outer side of the storage assembly, of the flexible blocking part according to the category of the wild animals detected by the multi-modal monitoring assembly, and therefore the safety of the wild animals is improved. The width of the inlet of the animal channel is adjusted, so that the animal channel meets the use requirements of different animals, and the utilization rate of the animal channel is increased.

The invention discloses a biological inspired camouflage image generation method, system and device, and relates to the field of camouflage images, and the method comprises the steps: obtaining a to-be-camouflage image; inputting the to-be-disguised image into an encoder to obtain stylized feature maps of multiple scales; the encoder is used for obtaining feature maps of multiple scales, calculating an average feature value of a background region in the feature maps, and rendering a foreground object based on the average feature value to obtain stylized feature maps of multiple scales; and inputting the stylized feature maps of the plurality of scales into a decoder for splicing to obtain a disguise image. According to the invention, by simulating the camouflage behavior of a chameleon in the nature and integrating the design concept of biological inspiration, the foreground object in the to-be-camouflaged image is matched with the background area by dynamically adjusting the color mode and the like of the foreground object, so that efficient visual camouflage is completed.

The application discloses a multi-waveband imitated vegetation composite camouflage material based on a multiple cooling mode and a preparation method thereof. The material comprises a high-hygroscopic camouflage base material, a high-reflective filler, phase change microcapsules and an additive. The high-hygroscopic camouflage base material not only serves as a matrix of various fillers and simulates the transpiration cooling effect of plants, but also, in combination with green pigments, can form a similar reflection spectrum to green plants; the high-reflective filler improves the reflectivity of the camouflage material in the visible light-near infrared-mid-infrared waveband, thereby realizing a radiation cooling effect; and the phase change microcapsules have a relatively high phase change enthalpy at normal temperature, thereby further reducing the overall temperature difference between the material and the environment background. The composite camouflage material is highly similar to green plants in the solar spectrum waveband, has a high correlation coefficient of 0.9778, can realize the same color and the same spectrum, and has a relatively low radiation temperature characteristic in the mid-infrared waveband. The camouflage material is expected to realize multi-waveband compatible camouflage in a vegetation background.

The present invention discloses an online supervision system and a supervision method based on construction safety, which relate to the field of safety supervision technology. The system comprises a protective shell, a loudspeaker is fixedly connected to the bottom of the protective shell, a warning light is fixedly connected to the top of the protective shell, and a solar panel is arranged above the warning light. By setting the exterior of the protective shell of the construction alarm in camouflage color and installing it in the woods outside the construction area, the system has strong concealment, which effectively prevents non-construction personnel from visiting the site and destroying it. Even if other construction alarms are destroyed and non-construction personnel successfully steal construction materials at night, the alarm can be sounded and evidence can be retained through such construction alarms hidden outside. The setting of the mimicry insect repellent mechanism utilizes folding leaves to simulate the process of bird wings spreading, flying and then recovering, which can effectively deter smaller animals and insects in the woods from nesting in the protective shell, so as to prevent damage to electronic components inside the protective shell, thereby increasing the service life of the construction alarm.

The invention discloses a camouflage color camouflage target detection method and system under a complex background and a medium, and belongs to the technical field of computer vision and target detection. Comprising the following steps: acquiring a camouflage painting target image, inputting the image into a pre-training detection model, and outputting target category and position information; the model is improved based on YOLOv8n, and an original backbone network is replaced by a Mobile ViT lightweight network in which a convolutional layer, an MV2 module and a Mobile ViT module are alternately connected; replacing the SPPF module at the tail end of the original trunk with a Focal Modulation module, and modulating features through a hierarchical aggregation and gating mechanism; a CPMS attention module is introduced behind two C2f modules of the original neck network, and multi-scale channel and space attention weighting is executed; a detection head with the scale of 160 * 160 is newly added to the detection head part and is used for detecting context sensing features fused with shallow details of the Neck part; according to the invention, high-precision and light-weight real-time detection of the camouflage target in a complex scene is realized.

The present invention relates to the technical field of camouflage sheds, specifically a retractable camouflage shed, including a towing mechanism, a roadbed laying mechanism installed on the towing mechanism, and a mechanism installed on the shed body. A top support plate and a bottom support plate that can push the gasket and the gasket respectively are movably installed on a slider at one end of the hydraulic sub-rod in the hydraulic rod away from the hydraulic mother pipe. Since the tie-in column is bolted to the outer end of the top plate of the load-bearing arm, the bottom end of the pressure-stabilizing end column will be positioned and clamped by the inner cavity of the tie-in column, and after the gasket and the gasket are stressed and stretched outward at the same time, the combined first umbrella frame, horizontal plate, second umbrella frame, clamp and vertical plate will expand in an umbrella shape, and multiple inclined plates, beams and reinforcing pads will cooperate with the port at the bottom of the camouflage outer layer to expand into a circular ground mat, so that its internal environment can form a relatively sealed state, while preventing mosquitoes, it can also provide protection on a more complex ground.

The utility model belongs to the technical field of camouflage net supporting, and discloses a state-adjustable camouflage net supporting device and a camouflage screen. The camouflage net supporting device comprises a supporting rod, a base and a net stretching plate assembly. Wherein the length of the supporting rod is adjustable, and the bottom end of the supporting rod is coaxially and detachably connected with the base; the net stretching plate assembly comprises a top cover, a plurality of net stretching plates and a switch, and a length-adjustable telescopic sleeve is arranged corresponding to each net stretching plate; wherein the top cover is coaxially and detachably connected to the top end of the supporting rod, and the multiple net stretching plates are distributed in the circumferential direction of the top cover at intervals; each net plate is rotationally connected to the top cover; one end of each telescopic sleeve is rotationally connected to the corresponding net stretching plate, and the other end of each telescopic sleeve is rotationally connected to a base of the switch; the switch can slide in the axial direction of the supporting rod and can be fixed to a preset position after sliding to the preset position. The utility model has the advantages of higher reliability and safety, convenience in disassembly and wide application scene.

The utility model relates to the technical field of thermal infrared camouflage, and discloses a smoke abatement and cooling camouflage device which comprises a shell, a filter bed and finned tubes. A flue gas inlet is formed in the first end wall of the shell, a flue gas outlet is formed in the second end wall of the shell, the flue gas inlet and the flue gas outlet are communicated through an inner cavity of the shell, and the multiple filter beds are arranged in the shell, so that high-temperature flue gas enters the inner cavity of the shell through the flue gas inlet, is filtered by the filter beds and then is discharged from the flue gas outlet; the finned pipes are connected end to end to form a cooling water pipeline, the cooling water pipeline is coiled in the filter bed, the initial end of the cooling water pipeline is connected with the water inlet pipe, and the tail end of the cooling water pipeline is connected with the waste heat utilization equipment through the water outlet pipe, so that high-temperature water generated after heat exchange between cooling water in the cooling water pipeline and the high-temperature flue gas is reused through the waste heat utilization equipment. The device effectively purifies the tail gas discharged by the diesel generating set, recycles the waste heat of the discharged smoke, reduces the thermal infrared intensity of the smoke, realizes visible light camouflage and thermal infrared camouflage, and is environment-friendly, energy-saving and efficient.

Surface pattern characterized in that a camouflage theme based on a combination of spots that exactly repeat the shape of the continents; execution in grayscale mode, namely in black, white, gray and its shades.

The invention relates to the technical field of camouflage material identification, in particular to a camouflage material identification method based on combination of deep learning and hyperspectrum, and the method comprises the steps: collecting hyperspectral data through a hyperspectral imaging system, carrying out the basic preprocessing of the collected hyperspectral data, carrying out the dimension reduction processing of the preprocessed hyperspectral data through a segmentation PCA algorithm, and carrying out the recognition of a camouflage material. A spectral imaging convolutional neural network model is constructed by adopting a method of combining deep learning and machine learning, accurate classification of camouflage materials is realized through multi-layer nonlinear processing, extracted characteristic wave band data is input into an SICN model, model training is carried out, performance evaluation is carried out on the model, and the accuracy of classification of the camouflage materials is improved. The hyperspectral data of the camouflage painting camouflage material to be recognized are input into the model, the model outputs probability distribution, the camouflage painting camouflage material is classified and recognized, and the spectral features of the camouflage painting camouflage material can be automatically learned by combining deep learning and the hyperspectral technology, so that rapid and accurate recognition is realized.

The invention relates to the technical field of camouflage, and discloses an intelligent production system of a camouflage film, which comprises a functional substrate with non-uniform physical characteristics; the reaction precursor layer is coated on the functional substrate and comprises an activating agent and an inhibiting agent; an energy field modulation unit, a process sensing unit, a control unit and a curing unit. The method comprises the following steps: providing a functional substrate and a reaction precursor layer; applying an initial energy field to trigger a self-organization process; monitoring the process in real time; predicting and determining a dynamic perturbation field through a digital twin model; applying the dynamic perturbation field for intervention; and finally, curing the formed pattern. By constructing the digital twin closed-loop regulation and control system based on real-time perception, predictive intervention on the self-organizing chemical reaction is realized, and the camouflage pattern which is highly related to the target environment, high in controllability and capable of adaptively fusing physical defects can be generated.

The utility model discloses a splicing type camouflage net, which relates to the technical field of camouflage nets and comprises a fixing frame, a plurality of flexible wires are mounted in the fixing frame, the fixing frame is formed by combining a fixing frame and a connecting shell, and the fixing frame is composed of a main body shell and a telescopic shell slidably connected in the main body shell. A fixing assembly used for fixing the position of the telescopic shell is arranged in the main body shell, a connecting groove used for being inserted into the connecting shell is formed in the fixing frame, and an elastic connecting piece used for bending is installed on the connecting shell; the camouflage net can be freely spliced, the camouflage net can be controlled to be folded to a proper angle to be laid through the elastic connecting pieces, meanwhile, certain bending and adjusting are carried out, and the size and the length of the camouflage net can be conveniently stretched.

The application discloses a dynamic camouflage wild animal passage monitoring device and method based on multi-modal perception in the technical field of wild animal research, which comprises a monitoring device and a passage shielding assembly arranged in an animal passage; the monitoring device comprises a dynamic simulation part and a multi-modal monitoring assembly arranged on the dynamic simulation part; the passage shielding assembly comprises a receiving assembly, a flexible blocking part arranged in the receiving assembly and a driving part for driving the receiving assembly to release the flexible blocking part; the multi-modal monitoring assembly is arranged in the inside of the dynamic simulation part, the multi-modal monitoring assembly is hidden by the dynamic simulation part, the interference of the monitoring device on animals is reduced, and the length of the flexible blocking part outside the receiving assembly is changed according to the wild animal category detected by the multi-modal monitoring assembly, so that the width of the animal passage entrance is adjusted, the animal passage is adapted to the use requirements of different animals, and the use rate of the animal passage is increased.