87results about How to "Reduce resolution" patented technology

The invention provides a three-dimensional geological modeling parameter optimization method and system based on deep learning, and relates to the field of geological prospecting and modeling.The method comprises the steps that drilling data and remote sensing data of a geological prospecting area are obtained, and multi-source geological data of three-dimensional geological modeling are formed; generating geological profile map data of the geological features according to the drilling data, generating landform profile map data corresponding to the topographic features according to the remote sensing data, and extracting boundary information of the geological exploration area; carrying out gridding processing on a three-dimensional space, dividing a geologic body of the geological exploration area into a plurality of small grid units, and defining geological modeling parameters; the preliminary three-dimensional geological grid model is constructed according to the geological profile map data, the landform contour map data and the boundary information, the size and the resolution of the grid are dynamically adjusted according to the geological change area, the three-dimensional geological model is obtained through optimization and adjustment, the modeling precision is improved, manual intervention is reduced, and computing resources are saved.

A non-contact three-dimensional displacement sensor disclosed by the present invention comprises a mechanical probe and a detection circuit, the mechanical probe part comprises a differential fixed block and an end cross beam, and two parallel inner side surfaces of the differential fixed block are fixedly provided with a first differential coil and a second differential coil respectively. A first target conductor located under the first differential coil is fixedly arranged on the top face of the suspension end of the cross beam, and a second target conductor located over the second differential coil is fixedly arranged on the bottom face of the suspension end of the cross beam. A pair of receiving coils Lx is fixedly arranged on the top face of the first target conductor or / and the bottom face of the second target conductor in the horizontal x direction, a pair of receiving coils Ly is fixedly arranged in the horizontal y direction, and the two pairs of receiving coils form a structure shaped like a Chinese character'jing '. The non-contact three-dimensional displacement sensor provided by the invention has better quasi-static resolution and displacement sensitivity in three directions, realizes decoupling in all directions, and has excellent performance in indexes such as linearity, measuring range and the like.

The invention discloses an enhanced photoacoustic microscopic imaging method based on physical degradation learning, which comprises the following steps of: on the basis of a photoacoustic imaging principle, degrading an image of a real photoacoustic microscopic image to the ground into a low-quality photoacoustic microscopic image through physical model degradation learning with variability. A high-quality photoacoustic microscopic imaging image is used as a target of supervised training, deep learning is used as a basis, the high-quality photoacoustic microscopic imaging image is input into a self-designed convolutional neural network with an attention mechanism for network training, the quality of a reconstructed image is evaluated until the reconstruction quality reaches the standard, and otherwise, the above steps are executed again. And inputting a real low-quality photoacoustic image into the trained neural network to obtain a reconstructed high-resolution photoacoustic image. According to the method, the resolution ratio and the contrast ratio of the deep photoacoustic microscopic image can be effectively improved, meanwhile, the photoacoustic microscopic image can be denoised, the dependence on the data set can be effectively reduced due to the fact that the data set is obtained through the physical model, and the reconstruction effect under different environments can be improved.

The invention relates to a touch all-in-one machine interaction control method and device, computer equipment and a medium, and the method comprises the steps: receiving image data through a preset video stream protocol, and obtaining a corresponding image data stream; acquiring a real-time network bandwidth condition, and then performing adaptive compression processing on the image data stream based on the real-time network bandwidth condition, and dynamically adjusting image transmission parameters; comprehensively analyzing the image data stream by using a deep learning algorithm, extracting operation characteristics of a user, generating an output control instruction based on the operation characteristics, and further controlling and adjusting image display and / or an equipment state; content information is extracted from the image data stream, and content analysis is carried out by adopting a corresponding content analysis strategy according to the content information, so that real-time display and / or storage are / is carried out according to the analyzed content; and obtaining a user interaction behavior, generating a corresponding adjustment feedback instruction based on the user interaction behavior, and performing personalized adjustment on the user interface. The method has the effect of improving the image transmission quality.

The invention provides an infrared weak and small target detection method and system, electronic equipment and a storage medium, and belongs to the technical field of target detection, and the method comprises the steps: improving a YOLOv8 model for the features of an infrared weak and small target image; and performing target detection on the infrared weak and small target image by using the improved YOLOv8 model. According to the invention, a stronger infrared weak and small target detection capability is provided for the YOLOv8 model, and the problem of target detection precision bottleneck of a traditional model in a complex scene is effectively solved.

The embodiment of the invention discloses an ultrasonic harmonic image generation method and device, electronic equipment and a medium. A specific embodiment of the method comprises the following steps: controlling an ultrasonic transmitting probe to transmit a low-frequency ultra-harmonic signal; controlling an ultrasonic receiving probe to receive the echo signal set to be detected; carrying out compressed sampling processing on the echo signal set to be detected and then carrying out signal reconstruction to obtain a reconstructed echo signal set; carrying out delay superposition processing on the reconstructed echo signal set to obtain a superposed echo signal; performing signal filtering decomposition processing on the superposed echo signals to obtain a super-harmonic signal set; performing envelope detection on the super-harmonic signal set to obtain an echo amplitude information set; and generating a target area ultrasonic harmonic image set, and sending the target area ultrasonic harmonic image set to a target terminal for display. According to the embodiment, by transmitting the low-frequency signal and receiving the ultra-harmonic signal, the attenuation of the received signal can be reduced, the signal penetration depth can be improved, the purity of the received signal can be improved, and the quality and the resolution of the generated ultrasonic harmonic image can be improved.

According to the building communication optical cable fault image processing method provided by the invention, on the basis of a U-Net model, by introducing a multi-scale aggregation pooling module MASP and a feature channel attention mechanism FCM, efficient capture of optical cable fault subtle differences and effective fusion of multi-scale features are realized, and the detection precision and efficiency are remarkably improved.

The invention discloses a remote sensing image semantic segmentation method, system and device and a medium, and the method comprises the following steps: obtaining remote sensing building data, and inputting the data into a pre-trained semantic segmentation network, so as to achieve the precise segmentation of a building in a remote sensing image. The semantic segmentation network includes an encoder module and a decoder module for adaptive context selection based on structure priori guidance. In the training stage, a special remote sensing building data set is utilized to train the network so as to optimize the segmentation performance of the network. According to the method, by introducing structure prior information and an adaptive context selection mechanism, the accuracy and efficiency of building segmentation in the remote sensing image are effectively improved, and the problem that remote sensing buildings are difficult to distinguish when pixel visual evidences are weak or the remote sensing buildings belong to small space structure objects is solved.

The invention discloses a tegillarca granosa blood hemoglobin concentration detection method and system based on image recognition, and belongs to the technical field of biology and computer crossover, and the method comprises the following steps: based on a tegillarca granosa blood sample, obtaining a sample image with a PyTorch tensor format; based on a convolutional neural network, the hemoglobin concentration of a sample image is marked, whether data is close to normal distribution is detected, the uniformly distributed data is used as a data set to train the convolutional neural network, a mean square error is used as a loss function, an Adam optimizer is used to carry out gradient descent optimization, and a detection model is constructed; and obtaining a blood sample of the tegillarca granosa to be detected, and identifying the hemoglobin concentration in the blood sample of the tegillarca granosa to be detected through the detection model. The method can quickly and conveniently complete determination and reduce the cost. The method can be used for rapidly and accurately measuring the hemoglobin concentration of tegillarca granosa, can also be used for measuring the hemoglobin concentration of other organisms, and has wide application and popularization values.

The invention belongs to the field of particulate matter optical sensing, and particularly relates to a particulate matter light scattering signal acquisition method and device, and the method comprises the steps: controlling scattered light with different scattering angles generated by particulate matter to be detected to reach a digital micromirror device DMD; enabling scattered light of different scattering angle combinations to reach a photoelectric detector by utilizing an on state and an off state of each micromirror in the DMD so as to realize multiple acquisition of scattered light signal intensity by the photoelectric detector and obtain a measurement signal vector S; wherein when a single micro-mirror is turned on, scattered light reaching the micro-mirror is focused on the photoelectric detector through the micro-mirror according to a preset light path, and when the single micro-mirror is turned off, the scattered light reaching the micro-mirror is diffused to an area outside the photoelectric detector; 1 norm minimization is carried out on X = (phi psi)-1S, and X and a sparse domain psi are determined; and performing inversion according to P = psi X to obtain the scattered light intensity P of different scattering angles generated by the particulate matter to be measured. According to the invention, the collection efficiency of scattered light intensity spatial distribution can be improved.

The invention discloses a remote control anti-delay video transmission method based on a future scene generation network. The remote control anti-delay video transmission method comprises the following steps: synchronizing time references of a remote control vehicle and a remote cockpit; original video stream data of the environment where the remote control vehicle is located is collected; adding a camera number and timestamp information into the original video stream data; preprocessing the decoded video stream data to generate a plurality of paths of future scenes; generating multiple paths of future scene video stream data; camera numbers and timestamp information are added to the multiple paths of future scene video streams; decrypting, decoding and distributing the data; performing packet loss and time delay detection on the original video stream data, and performing visual quality optimization on the future scene video stream data; and a complete and continuous video stream is formed. According to the method, the delay in video transmission is greatly reduced, and the video content at the future moment can be predicted and generated in advance in future scene generation, so that the video blank possibly occurring due to transmission delay or data packet loss is effectively filled.

The invention relates to the technical field of thoracoscope lens cleaning, in particular to a thoracoscope lens cleaning device which comprises a thoracoscope used for collecting images of a surgical site in real time; the supporting tube is used for supporting the thoracoscope to enter a surgical incision; the cleaning mechanism comprises a decontamination cover, a cleaning pipe and a cleaning base and is used for cleaning the lens; the negative pressure suction mechanism comprises a smoke suction pipe, a plurality of smoke suction holes and a siphon opening; the data acquisition module is used for acquiring an incident light spectrum and a transmission light spectrum of the lens in real time; and the control mechanism is used for determining the resolution of the image so as to judge whether to start to clean the lens and the qualification of the cleaning effect of the lens, determining the distribution mode and the pollution index of pollutants based on the comparison result of the incident light spectrum and the transmitted light spectrum so as to determine the cleaning mode, and determining the cleaning parameters of the lens based on the pollution index. The cleaning efficiency of the thoracoscope lens is improved.

The invention provides a three-dimensional Gaussian splash algorithm training method and device, equipment and a medium, and belongs to the technical field of image processing, and the method comprises the steps: obtaining an image data set, carrying out the down-sampling of an image in the image data set based on at least one sampling rate, and determining a target image data set; determining a three-dimensional Gaussian distribution model according to the image data set; performing multi-round iterative training on the three-dimensional Gaussian distribution model according to the target image data set and the two-dimensional filter; each round of iterative training determines an anti-aliasing filter according to the target image data set and a preset two-dimensional filter; and filtering the three-dimensional Gaussian distribution model according to an anti-aliasing filter, and training the filtered model. According to the method, iterative training is carried out by using a low-resolution image instead of an original image, the calculation complexity is reduced, training acceleration is realized, the three-dimensional Gaussian distribution model is filtered through the anti-aliasing filter, the aliasing effect is overcome, and the rendering quality is ensured while the training time is shortened.

The embodiment of the invention discloses a sleep monitoring method and device, electronic equipment and a storage medium. The sleep monitoring method comprises the steps of receiving a reflection signal sent by a millimeter wave radar monitoring user, processing the reflection signal, generating multi-frame point cloud information, and determining sleep information of the user according to the change degree of the point cloud information in a monitoring area. According to the embodiment of the invention, the reflected signal of the position of the target user is acquired through the millimeter-wave radar, so that the user does not need to wear any equipment, the use is more comfortable, the reflected signal acquired by the millimeter-wave radar generates the point cloud information accurately representing the user state, the sleep condition of the user is accurately identified, and the user experience is improved. Therefore, the sleep health condition of the user can be accurately evaluated, and the user experience is further improved.

The invention discloses a PCB (Printed Circuit Board) defect detection method and equipment for an edge MCU (Microprogrammed Control Unit) and a storage medium, and belongs to the technical field of PCB defect detection. The method comprises the following steps: acquiring a PCB image; carrying out down-sampling optimization on the YOLOv10 model by adopting deformable convolution and separation characteristics respectively; and the optimized YOLOv10 model is adopted to detect the defects of the PCB. According to the method, the requirements of the PCB defect detection model for storage and computing resources can be remarkably reduced, high reasoning speed and accuracy can be kept, the problem of real-time data processing in a weak network environment is solved, and optimization of the response speed and safety of a whole industrial internet system is facilitated.

The invention provides a safety and reliability evaluation system for an intelligent sensing system, which can be applied to the technical field of automatic driving evaluation and comprises a data acquisition module for acquiring initial fault-free sensor data and transmitting the initial fault-free sensor data to the intelligent sensing system; the fault injection module is used for performing fault data injection on the initial fault-free sensor data to obtain sensor fault data and transmitting the sensor fault data to the intelligent sensing system; the result analysis module is used for receiving and analyzing the test data of the intelligent sensing system to obtain a test result; and the interactive module is used for controlling the work of the data acquisition module, the fault injection module and the result analysis module. Fault information of different formats is generated according to data formats of different types of sensors, and real-time sensor data reading, fault signal injection and data recording and analysis can be achieved. The invention further provides a device for evaluating the safety and reliability of the intelligent sensing system.

The invention discloses a preparation method and application of a wide-range flexible capacitive pressure sensor, and the method comprises the steps: obtaining the parameter information of a high-modulus dielectric material based on a test result of a single-layer flexible pressure sensor; simulation analysis is carried out based on the high-modulus medium parameter information, and the optimal modulus gradient range between the media is determined; according to the obtained optimal modulus gradient range data, low-modulus dielectric material parameters are determined; respectively selecting low-modulus polymer powder and high-modulus polymer powder according to the parameters of the low-modulus dielectric material and the parameters of the high-modulus dielectric material to prepare a low-modulus dielectric film and a high-modulus dielectric film; and assembling the low-modulus dielectric film, the high-modulus dielectric film and the flexible electrode according to a sandwich structure, and leading out a lead to obtain the double-layer flexible pressure sensor. According to the invention, smooth conversion of force between two kinds of modulus media is ensured, and the phenomenon of force conduction failure caused by modulus mismatch between the media is avoided.

The embodiment of the invention belongs to the field of artificial intelligence, and relates to a face-scanning payment method, which comprises the following steps: when a face-scanning payment request is received, obtaining a first image of a customer through a POS terminal; performing image reconstruction on the first image through an image super-resolution reconstruction algorithm to obtain a high-resolution second image; inputting the second image into a face detection model to obtain a face region image in the second image; performing feature extraction on the face region image through a local binary pattern algorithm to obtain a face feature image; performing face recognition in a preset face feature database based on the face feature image to obtain an identity verification result of the face scanning payment request; and when the identity verification result shows that the client identity is legal, responding to the face scanning payment request to complete the face scanning payment operation. The invention further provides a face-scanning payment device, computer equipment and a storage medium. According to the invention, the accuracy of face-scanning payment of the POS terminal is improved.

The invention relates to the technical field of grid map construction, and discloses a visual robot grid map construction method comprising the following steps: 1, grid granularity size selection; step 2, grid order marking; step 3, obstacle grid adjustment; step 4, establishing a grid map; in the third step, feature extraction is carried out on obstacles, images and videos in the surrounding environment are obtained through a visual sensor assembly, and then the images and the videos are processed and analyzed to extract useful information, so that information of the obstacles in the grids can be sensed. According to the method, in the grid map construction process, the particle sizes of a plurality of grids are set to be large, for example, when obstacles exist in part of grids and the particle sizes of the grids are adjusted, the grids with the obstacles are subdivided into a plurality of 10 * 10 small grids, so that grid map construction can be carried out more accurately; and the subsequent decision-making speed is improved while the information storage amount of grid map construction is reduced.

The invention provides a high-throughput wide-area Raman spectrum system for bacterial colony detection and an implementation method. The high-throughput wide-area Raman spectrum system comprises a laser and light spot adjusting module, a Raman imaging module and a detection module, the spatial heterogeneity problem of a bacterial colony sample is solved by setting a light spot adjustment scheme, a bacterial colony detection scheme and a high-throughput wide-area Raman scheme, so that the spectral repeatability of the bacterial colony sample is better, and establishment of a spectral database and identification of the bacterial colony sample are facilitated; the system is different from a common wide-area Raman system, has high throughput and high resolution, is more suitable for analysis and identification of biological samples, meets the resolution requirements of analysis and identification of various biological samples, and is scientific and reasonable in overall structural design, safe and reliable in implementation process and suitable for popularization and application.

The application is suitable for the technical field of laser radar, and provides a laser radar, which comprises a light source, a transmitting module, a receiving module, a transmission module, a detecting module and a signal processing module; the transmission module comprises a first coupler and a plurality of transmission media, wherein the local oscillator beam is transmitted to at least one transmission medium for delay processing through the first coupler, and the delay amount of the at least one transmission medium is different from that of the other transmission media; the detecting module is connected with the transmission module and the receiving module respectively, is used for receiving the echo signal and the delayed local oscillator beam output by the transmission module, performing beat frequency processing on the echo signal and the delayed local oscillator beam to obtain a beat frequency signal, and converting the beat frequency signal into a detecting electric signal; and the signal processing module is electrically connected with the detecting module, is used for determining a to-be-measured object according to the detecting electric signal. The laser radar provided by the application can improve the strength of the detection signal in long-distance measurement.

The invention provides a clinical image-based cancellous bone mechanical property prediction method and system, and the method comprises the steps: carrying out the medical CT scanning of a plurality of prepared cancellous bone cubic samples, and obtaining the medical CT image data; carrying out segmentation processing on the medical CT image data, extracting a sample area and calculating an average HU value of a sample CT image; performing a uniaxial compression experiment on the sample, accurately measuring strain by using a digital image correlation method, and further obtaining a stress-strain curve to calculate an elastic modulus and compression strength; according to the average HU value and the corresponding mechanical index data, a linear relation model of the average HU value of the cancellous bone CT image and the elastic modulus and the compression strength of the cancellous bone is established, and then a personalized treatment scheme is formulated according to the characteristics of the cancellous bone of the patient. The method has high universality and accuracy, and the elastic modulus and the compression strength of the cancellous bone sample can be accurately measured.

The application provides a method for converting a 32-bit face depth map into an 8-bit depth map for live body detection, and the method comprises the following steps: S1, collecting a TOF near-infrared image and a depth map; and processing the 8-bit near-infrared image through S2 and the 32-bit depth map through S4; S2, performing face detection on the near-infrared image; S3, expanding the face detection frame; S4, converting the 32-bit depth map into an 8-bit depth map; and S5, performing face frame cropping. According to the method, the pixel information of the face in the 32-bit depth map can be obtained through face detection on the original 8-bit near-infrared image and the position of the face frame, the pixels of the face are reserved by operating the pixels, the 8-bit depth map of the face is finally converted, the important pixel information is not lost in the depth map, and the color depth difference is maximized.

The application relates to the technical field of industrial lens, and discloses a low-distortion macro lens, which comprises a mechanical device and an optical module installed in the mechanical device; the optical module is sequentially provided with a first lens group T1, a diaphragm, a second lens group T2 and a third lens group T3 from an object side to an image side; the first lens group T1 has positive refractive power, the focal length of the first lens group T1 is f T1 ; the second lens group T2 has negative refractive power, the focal length of the second lens group T2 is f T2 ; the third lens group T3 has positive refractive power, the focal length of the third lens group T3 is f T3 ; the total focal length of the optical module is f, 0.50<|f T1 / f|<0.85, 3.00<|f T2 / f|<4.00, 1.60<|f T3 / f|<2.50. The low-distortion macro lens realizes an optical system of a high-resolution low-distortion macro lens with a focal length of 50 mm, the maximum optical distortion of a full field of view is lower than 0.14%, and the problem of poor close-range imaging effect of an industrial macro lens is solved.

This invention discloses a time amplifier with self-cancelling quantization noise and a fractional-frequency division phase-locked loop (PLL), belonging to the field of frequency synthesis technology. The self-cancelling quantization noise time amplifier of this invention is based on a resistor-discharge type time amplifier. A control signal containing quantization error information is output through a slow-slope control signal generator to control the slow-discharge path of the time amplifier. This achieves K-fold time error amplification while simultaneously canceling fractional-frequency division quantization noise. The PLL architecture built based on this amplifier eliminates the need for a large-delay loop digital-to-time converter. The phase noise of the slow-slope control signal generator is suppressed by the time amplification gain, breaking the performance trade-off of traditional architectures. This invention achieves a low-noise, low-power, and high-linearity fractional-frequency division PLL design without degrading the performance of the time amplification phase detector.

The invention provides a video data multi-scale down-sampling method and device based on an FPGA. The method comprises the following steps: counting line and field synchronization signals of received image data; according to the counting result of the row-field synchronizing signals, generating down-sampled row-field marking signals; enabling originally received image data to correspond to the generated down-sampling line field marking signal, carrying out image down-sampling, and writing the image into an FIFO (First In First Out); integrating the down-sampled image data into continuous line field image data by using FIFO (First In First Out), and outputting the continuous line field image data; and generating a new line field synchronization signal according to the down-sampled image data, aligning the new line field synchronization signal with the down-sampled line field image data, and synchronously outputting the line field synchronization signal and the down-sampled line field image data. According to the invention, multi-scale down-sampling of the video data based on the FPGA can be realized, down-sampling is carried out on a high-resolution image, the image resolution is reduced, the pressure of FPGA image processing is reduced, the sampling scale of down-sampling of the video data is variable, and setting of different sampling scales can be carried out according to actual needs.

The application belongs to the technical field of image processing, and relates to a light field microscopic image processing method based on a microlens array. The application establishes an optical path model of a light field microscope imaging system based on the microlens array; according to size data of a target microscope, the microlens array is designed, and all parameters of the optical path model are determined; according to optical parameters of the optical path model, various parameters required for high-resolution image reconstruction are designed; the microlens array is manufactured according to the determined microlens array parameters; the manufactured microlens array is integrated into a traditional microscope to assemble the light field microscope; the assembled light field microscope is used to collect an image array, the image array is divided into multiple sub-images, and fusion processing is performed to obtain a final target image. Without improving the hardware complexity, the application significantly enhances the contrast and definition of the image, and effectively solves the technical problems of low spatial resolution and low signal-to-noise ratio of the existing light field microscope.

The invention relates to the technical field of image processing, and particularly discloses an intelligent security and protection monitoring method and system for an intelligent building based on the Internet of Things, and the method comprises the following steps: S1, carrying out the framing processing of a monitoring video, obtaining a monitoring video frame sequence, and determining a target region in the monitoring video; s2, determining a playing window of the target video on the reference plane, and marking a target area in the playing window; moving the reference line along the target direction, and recording the area proportion every preset duration; correcting the area proportion according to the proportion of the target area in the playing window to obtain a correction proportion, obtaining a target sequence based on the correction proportion, and determining a comparison value; s3, determining a screening sequence according to the comparison value, and counting a comparison number; and determining an adjustment window based on the comparison number, and preprocessing the target area in the adjustment window. According to the invention, the resolution of the overlapping region is reduced, the video data volume is reduced, and the normal operation of the monitoring system is ensured.

This invention discloses a method, apparatus, and product for detecting small targets in drone aerial photography, relating to the field of computer vision, to improve the detection performance and parameter efficiency of small targets in complex background scenes. In the visual feature extraction process, the aerial image is downsampled hierarchically using a spatial-to-channel rearrangement approach to obtain multi-layered downsampled feature maps. A multi-branch convolutional structure is used to enhance the receptive field of each downsampled feature map, resulting in multi-scale visual features. In the text semantic enhancement process, at least one cue vector most relevant to the aerial image is selected from a cue vector library. The cue vector is semantically enhanced to obtain enhanced semantic features, which are then fused cross-modally with the multi-scale visual features. A detection head is then used to identify the target based on the cross-modal fused features. This invention improves the accuracy, robustness, and generalization ability of small target detection in complex environments.

The present invention provides a measurement system for scattered target points of shaped springs based on a single camera, comprising a measured shaped spring, a light source, a reflector, a microscope lens and an industrial camera; the measurement light propagation path of the system is that the light source emits the measurement light, the measured shaped spring reflects the measurement light, the measurement light passes through the reflector and then passes through the microscope lens, and then enters the industrial camera. The present invention can economically and effectively perform accurate visual measurement of the geometric dimensions of scattered target points of shaped springs. On the one hand, it overcomes the disadvantage that the pixels of the image are not fully utilized in the measurement scheme of a single telecentric lens camera, and also solves the mechanical interference problem of the measurement scheme using only multiple microscope lens cameras; on the other hand, compared with a single microscope lens plus a precise motion platform, the present invention can significantly reduce costs while ensuring measurement accuracy.