15results about How to "Good test result" patented technology

The invention discloses a rotating target detection method and system based on phase shift coding and directional feature enhancement, and belongs to the technical field of image recognition. The method comprises the steps of obtaining an image data set and performing enhancement processing; inputting into an ARC network to extract features of a target in any direction; performing adaptive feature optimization on the features through an EMA module, encoding optimized feature direction information by using ARF, generating direction sensitive features for bounding box regression, and extracting directional features for classification through a rotation invariant pooling layer; linearly mapping the angle of the rotating frame into a phase of a corresponding frequency by using a phase shift encoder, predicting a phase shift cosine value, and decoding to obtain a final phase; and finally, removing overlapped detection frames by applying a non-maximum suppression method, and reserving an optimal detection result. Wherein a Pareto optimization theory is used for adaptively weighting a loss item, model parameters are updated through a multi-gradient descent algorithm, and the obtained final model can improve the accuracy and efficiency of detection of a target containing a multi-directional small object.

The application relates to the technical field of biological medicine, in particular to a medicine for combined treatment of tumors by using a recombinant oncolytic virus and a small-molecule anticancer drug. The method comprises the following steps: combined treatment of tumors by using a recombinant oncolytic virus and a small-molecule anticancer drug; the small-molecule anticancer drug is a small-molecule anticancer drug targeting Parp; the recombinant oncolytic virus comprises M protein, G protein, N protein, P protein and L protein after site-directed mutagenesis. The application can kill tumor cells by using a recombinant oncolytic virus and a small-molecule anticancer drug, and the curative effect is 1+1>2.

The present invention relates to the field of deep learning and computer vision technology, and in particular to a high-resolution face anti-spoofing method and system for effectively learning face details, wherein the method comprises the following steps: S1, through an adaptive feature enhancement network, feature weighting is performed on a face image and its corresponding depth map and reflective map; S2, through a diagonal feature fusion network, feature extraction is performed on the feature-weighted face image to obtain a discriminative face local detail feature map; S3, through a deception area detection network, the face local detail feature map is amplified, and face deception feature areas are extracted in a high-dimensional space; S4, through a fully connected layer and a SoftMax classifier, the face deception feature area is processed to determine whether the face is a deception face. The present invention can improve the recognition of face deception areas and further improve the efficiency of face liveness detection.

The utility model relates to a rubber air bag pressure detection platform of an air shield dam, and belongs to the technical field of inspection and testing of water conservancy and hydropower facilities. The rubber air bag pressure detection platform is a device for detecting and testing pressure resistance and air tightness of the rubber air bag; the device comprises an inspection device, an air supply system, an exhaust system and a rubber air bag of the air shield dam. The inspection device is composed of an upper clamp, a lower clamp, a rotating shaft of the upper clamp and the lower clamp, clamp supports and fastening bolts. The upper clamp and the lower clamp are semi-I-shaped, are formed by two steel plates with the thickness of 50mm through laser cutting, and are respectively welded with the upper steel of the full-length fillet; and the air supply system and the exhaust system are respectively connected through a three-way valve connected with an air tap of the rubber air bag.

The invention relates to the technical field of detection, in particular to a building curtain wall flatness detection device. According to the technical scheme, the detection assembly comprises a rotating part, a circular plate, a middle part, a detection plate, a roller set, a deviation plate and a buffer air spring, one end of the rotating part is connected with the circular plate, the middle part is fixedly installed on the side, away from the rotating part, of the circular plate, the middle part is of a U-shaped structure, and the detection plate is rotationally connected into an opening of the middle part; and a roller group is connected in the detection plate in a rolling manner. The round plate drives the middle piece and the detection plate to rotate on the wall surface, so that the roller set rotates on the wall surface, when the wall surface is uneven, the detection plate inclines along with the wall surface, the deviation plate is used for extension, and when the deviation plate deviates slightly, the displacement distance of the end part is large, so that the flatness detection result of the wall surface is more obvious, and the detection efficiency is improved. The flatness is detected through rotation of the detection plate, and the detection efficiency is improved.

The invention discloses a wide linear minimum mean square error-based MIMO-OCDM detector detection method and system, and the method comprises the steps: demodulating a signal received by a receiving antenna based on discrete Fresnel change, and obtaining an observation signal in an MIMO-OCDM communication system. Constructing a WL-MMSE detector, and detecting the observation signal by using the detector to obtain a corresponding estimation signal; and processing the estimated signal by using an LAS algorithm, and searching an optimal solution in a neighborhood of the estimated signal through successive symbol flipping and likelihood function optimization to complete detection. According to the invention, high signal-to-noise ratio gain is realized, the actual operation complexity is greatly reduced, and the method is more suitable for high-speed, low-delay and high-capacity 6G application scenarios.

The invention relates to the technical field of conductivity testing, and provides a device and a method for testing the conductivity of a filler for a polymer, which can improve the accuracy and the reliability of testing by integrating data of a plurality of testing methods and constructing a conductivity testing model knowledge base. The device comprises five units, namely, a data acquisition unit, a data processing unit, a model matching unit, a result generation unit and an optimization unit, and a full-automatic conductivity testing process is realized. The change rule of the conductivity of the filler under different conditions can be deeply understood by controlling a variable dynamic test. According to the invention, personal errors are reduced, the test efficiency is improved, the comparability of test data among different laboratories or research institutions is promoted, and powerful support is provided for research and application of the conductivity of the filler for the polymer. The device and the method have a wide application prospect.

This application relates to the field of biomedicine, specifically to a drug for the combined treatment of tumors using a recombinant oncolytic virus and a small-molecule anticancer drug. The specific steps include: treating tumors using a combination of a recombinant oncolytic virus and a small-molecule anticancer drug; the small-molecule anticancer drug is a BTK-targeting small-molecule anticancer drug; the recombinant oncolytic virus includes site-directed mutated M, G, N, P, and L proteins. This application utilizes the combined attack and killing of tumor cells by a recombinant oncolytic virus and a small-molecule anticancer drug, achieving a therapeutic effect greater than the sum of its parts (1+1>2).

The utility model discloses a cloth inspecting machine, relates to the technical field of textile equipment, and adopts the technical scheme that the cloth inspecting machine comprises a cloth inspecting machine body, a cloth inspecting table and a winding device are arranged on the cloth inspecting machine body, a cloth inspecting area is arranged on the cloth inspecting table, a plurality of cloth inspecting lamps which are distributed at equal intervals are arranged in the cloth inspecting area, and a partition plate is fixedly connected between every two adjacent cloth inspecting lamps; limiting pieces which are symmetrically distributed are arranged on the portions, located on the two sides of the cloth inspecting area, of the cloth inspecting table, and the limiting pieces are arranged in a telescopic adjusting mode. According to the cloth inspecting device, the cloth inspecting lamps are arranged in the cloth inspecting area, the cloth inspecting lamps in the corresponding area can be turned on according to the size of cloth needing to be inspected, energy waste is reduced, the limiting pieces are arranged on the two sides of the cloth inspecting area, the cloth can be limited to a certain degree in the cloth inspecting process, left-right sliding of the cloth is avoided, and the cloth inspecting efficiency is improved. The cloth inspecting machine has a better inspecting effect, meanwhile, the limiting pieces can also play a role in shielding redundant light sources on the two sides of the cloth, and operators can observe the cloth conveniently.

The utility model relates to the field of textile printing measuring equipment, and provides a textile deformation measuring instrument, which comprises a bottom plate, one end above the bottom plate is fixedly connected with a first base, the other end above the bottom plate is provided with a chute, the other end above the bottom plate is movably connected with a second base, and the second base is provided with a sliding groove. A sensor body is installed at the top end of the second base, pulleys are arranged at the bottom end of the second base, a distance adjusting assembly is arranged below the inner wall of the first base, and clamping assemblies are arranged above the inner walls of the first base and the second base. According to the technical scheme, the sample needing to be measured is clamped through the clamping assembly, and the problem that the comparability and accuracy of a test result are poor possibly due to the fact that the position of the sample deviates during measurement is solved.

The present invention discloses an Internet-based blood tumor marker precision and efficient analysis system and a method for using the same, and belongs to the field of medical device technology. An Internet-based blood tumor marker precision and efficient analysis system includes a detection box, support seats are provided on both sides of the top of the detection box, a stirring drum is movably connected between the two support seats, the top and bottom of the stirring drum are respectively connected to a liquid inlet pipe and a liquid outlet pipe, a stirring mechanism is provided on the liquid inlet pipe, a moving rod for driving the stirring mechanism is movably connected in the liquid inlet pipe, a sealing piston is slidably connected to the inner wall of the liquid outlet pipe, and the sealing piston is fixedly connected to the moving rod; the present invention squeezes the detection card into a concave shape so that the dripping liquid will not flow out and can be completely absorbed by the detection card, the detection result is more obvious, and it will not drip into the placement slot, and there is no need to increase the subsequent cleaning work, and at the same time, the tumor markers in the blood are efficiently stirred, and the stirring effect is improved, which is convenient for increasing the accuracy of the detection effect.

The application discloses a kind of construction method and application of intelligent signal detection model, belong to mobile communication technical field;The application constructs a new type of graph neural network containing aggregation update module, the global feature of each node is aggregated with its neighbor node information, so as to extract the hidden feature state of data according to the feature information of neighbor node and itself, and then realize the effective capture of the feature of communication signal such non-euclidean space data;In addition, after each aggregation is completed, the intermediate feature obtained by aggregation is further extracted in the application Feature, to improve the expression ability of intermediate feature, so that the designed graph neural network has more efficient information interaction ability and more powerful relationship reasoning ability, greatly improve the precision of signal detection.The application effectively solves the problem that the performance and computational complexity of traditional signal detection method are difficult to balance, and has strong robustness to channel estimation error.

This invention provides a road obstacle detection method based on an improved version of YOLOv4. Improvements have been made to YOLOv4, making the network more lightweight, enhancing its robustness and ability to handle complex environments, and enabling more accurate obstacle detection, thus improving the detection performance for small targets. The method includes: using the lightweight MobileNetV3 as the backbone network to provide feature extraction operations; employing a linear bottleneck inverse residual structure combined with depthwise separable convolution and SE attention mechanisms; improving the feature fusion network; using an efficient channel attention module in the model, which avoids dimensionality reduction and effectively captures cross-channel interaction information, maintaining the number of channel layers after global pooling, and using a non-dimensionality-reduced local cross-channel interaction strategy; focusing on the relationships between different channels through band matrix learning, selectively learning different information in different channels; and proposing a new DBR module, which consists of DepthwiseConv2D, BatchNormalization, and ReLU6.

This application relates to the field of biomedicine, specifically to a drug for the combined treatment of tumors using a recombinant oncolytic virus and a small-molecule anticancer drug. The specific steps include: treating tumors using a combination of a recombinant oncolytic virus and a small-molecule anticancer drug; the small-molecule anticancer drug is a CDK4 / 6-targeting small-molecule anticancer drug; the recombinant oncolytic virus includes site-directedly mutated M, G, N, P, and L proteins. This application utilizes the combined attack and killing of tumor cells by a recombinant oncolytic virus and a small-molecule anticancer drug, achieving a therapeutic effect greater than the sum of its parts (1+1>2).