19results about How to "Reduce network parameters" patented technology

The invention relates to the technical field of prostate MRI (Magnetic Resonance Imaging) image segmentation, in particular to a prostate MRI image segmentation method based on Mamb-Unet, which realizes effective fusion of features of different scales between an encoder and a decoder, and fuses multi-scale information on a space axis by using a space attention bridge segment to fuse fusion features of different levels from the encoder, so that the segmentation accuracy of the prostate MRI image is improved. And then global and local information is captured through 1D convolution and full connection in the channel attention bridge segment, so that a channel attention map is generated, channel information is better integrated, and multi-level different-scale feature fusion is realized.

The invention provides a subway tumble detection method and system based on cloud edge collaboration and an X-CLIP model, and relates to the technical field of subway tumble detection, and the method comprises the steps: obtaining a real-time video stream in a subway, and analyzing the video stream data into frame image data and video stream data; inputting the frame image data into the YOLOv8n-c3xghost neural network model to obtain all human body target area detection frames; performing identification detection on the human body target area detection frame through a BoT-SORT target tracking algorithm to obtain candidate images of abnormal behaviors; inputting the candidate image of the abnormal behavior into an X-CLIP model, outputting an abnormal region fragment classification result of the candidate image, identifying an abnormal region fragment again, determining the falling behavior according to the situation that the falling probability is greater than a set threshold value, outputting a position coordinate where the falling behavior occurs and a corresponding original image, and giving an alarm; wherein the model is deployed at the cloud end and trained, the model sinks to the edge side after being compressed and packaged, and real-time video frame detection is realized at the edge side.

The invention provides a multi-task-based power transmission line stockbridge damper defect detection method, which comprises the following steps: acquiring an original power transmission line stockbridge damper image, and generating a data sample close to a real fault sample; a multi-task detection network model is established, a classification loss function and a segmentation loss function are constructed for the multi-task detection network model, and the established multi-task detection network model comprises a shared feature layer, classification branches and segmentation branches; based on a data sample, training the multi-task detection network model by adopting an alternate training mode to obtain a trained multi-task detection network model, and in the training process, adaptively adjusting weight parameters by adopting an adaptive dynamic weighting strategy; and inputting a to-be-detected image of the vibration damper of the power transmission line into the trained multi-task detection network model to obtain segmentation positioning and defect detection results of the vibration damper of the power transmission line. According to the invention, the problem of poor robustness and detection performance in the prior art can be solved.

The invention provides a grounding grid corrosion defect identification method based on an edge enhancement model, a medium and equipment, and relates to the technical field of grounding grid corrosion defect identification. On the basis of a traditional LeNet-5 network structure, two channels with different scales are connected in parallel to form a three-channel parallel module, macroscopic and microscopic information on different scales can be extracted, and the mutual relation between the macroscopic and microscopic information can be captured through cross-scale convolution, so that the information between the macroscopic level and the microscopic level is fused, and the information fusion efficiency is improved. The expression capability of the network on corrosion defects in different shapes is improved; by adding the edge extraction module, a traditional full connection layer is improved, network parameters can be greatly reduced, the overfitting problem is effectively avoided, meanwhile, the network structure is simplified, and the identification accuracy of the corrosion defect of the grounding grid is effectively improved.

The invention discloses a deep learning-based intelligent identification method for animal images of infrared cameras along a railway. A trunk module, a neck module and a detection head module which are connected in sequence are included; the trunk module comprises an input layer, a spatial pyramid pooling layer, a position space attention layer, two spatial channel decoupling down-sampling layers, three Conv layers and C2f layers, a C2fCIB layer and an output layer, and the trunk module outputs feature maps of five scales; the neck module is composed of a path from bottom to top and a path from top to bottom and comprises an input layer, a space channel decoupling lower sampling layer, a Conv layer, a C2f layer, two dynamic upper sampling layers, three C2fCIB layers and an output layer. The detection head module includes a pair of heads and a pair of multiple heads, each head including regression and classification branches. According to the identification method, a dynamic up-sampling layer DySample based on point sampling is adopted, the characterization capability of the feature map is improved, the model is helped to recover the feature resolution, the classification and positioning accuracy is improved, and therefore the identification precision is improved.

The invention discloses a lightweight human body posture estimation method based on Lite GS-HRNet, and belongs to the technical field of computer vision posture estimation, and the method comprises the following steps: downloading a COCO data set; performing data preprocessing on the color human body images in the data set; the preprocessed image is input into a Lite GS-HRNet network for training; through the trained Lite GS-HRNet network, bone joint point information of the human body is output; according to the method, a Ghost module and a SimAM attention mechanism are combined, network parameters and calculation complexity are effectively reduced, high-precision attitude estimation is kept while real-time performance is ensured, the applicability of augmented reality and intelligent monitoring conditions needing real-time processing is improved, and the method is suitable for popularization and application. And the use accuracy of the mobile equipment, the edge calculation and the real-time processing application condition is improved.

The application discloses a speech separation method and device based on a multi-channel full convolution time domain network, comprising the following steps: (1) obtaining a plurality of noisy mixed multi-channel speech signals containing different sound sources, noise and reverberation, and taking corresponding pure single-channel speech signals as labels to form a training data set; (2) establishing a multi-channel full convolution time domain network, wherein the multi-channel full convolution time domain network comprises an encoder, a separator, a point multiplication module and a decoder, and the parameters of the encoder are fixed Gammatone filter coefficients; (3) inputting the training data set into the multi-channel full convolution time domain network for training; and (4) inputting a noisy mixed multi-channel speech signal to be separated into the multi-channel full convolution time domain network to obtain a pure single-channel speech signal after sound source separation. The application has better separation effect.

The invention provides a video tampering detection method and system based on multi-scale features and a hybrid 3D network, and relates to the technical field of computer vision, and the method comprises the steps: obtaining a to-be-detected video frame, and carrying out the preprocessing of the to-be-detected video frame; feature maps of different scales are obtained through sampling processing; inputting the feature maps of different scales into a multi-scale internal cascade network, and extracting shallow features, time features and spatial features by using a shallow feature extraction module, a time feature extraction module and a spatial feature extraction module respectively; carrying out feature fusion on the shallow features, the time features and the spatial features by adopting an EMA fusion module with multi-scale features to obtain fused features; performing video frame detection by using the fused features to obtain a video tampering detection result; wherein a hybrid three-dimensional convolution structure is introduced into the time feature extraction module. According to the method, whether the video frame is subjected to illegally inserted tampering operation or not can be effectively identified.

The present application provides a method and system for real-time identification of natural earthquakes at sea using small sample lightweight deep learning, the method comprising: inputting the sound pressure signal recorded by a snorkeling marine seismograph into multiple trained lightweight deep learning networks, and outputting whether a natural earthquake at sea has occurred; for the output results of multiple networks, a voting mechanism is adopted, and the voting result is used as the final recognition result; the training of the lightweight deep learning network comprises: preprocessing the sound pressure signal recorded by the snorkeling marine seismograph; using a supervised learning method to perform multiple model training; the preprocessing comprises: uniformly sampling the data to a fixed frequency and a set length, and performing a 0.05-3Hz bandpass filtering process on the data. The advantages of the present application are: the ability to accurately identify natural earthquake signals in complex background noise; and the ability to be deployed on resource-constrained embedded devices and mobile devices with high requirements for real-time processing.

The present invention discloses a deep learning wafer solder joint detection method based on improved YOLOV5, the method comprising: 1) preparing a wafer solder joint data set; 2) constructing an attention mechanism module CCANET; 3) constructing a YOLOV5 network incorporating the attention mechanism; 4) introducing a Ghost module; 5) training the improved YOLOV5 network. The present invention improves the accuracy of wafer solder joint detection while using fewer network parameters. Under the same conditions, not only more wafer solder joints can be detected, but also blocked wafer solder joints can be detected.

The application discloses a pyramid type time convolution network structure and a diagnosis method for real-time bearing fault diagnosis, relates to the technical field of mechanical part fault diagnosis, and is composed of four parts of data input, network segment 1, network segment 2 and result output. The data input part is used for preprocessing original data and transmitting the original data to the network segment 1, the network training is performed through the connection mode of neurons in the network segment 1, and the output prediction sequence is transmitted to the network segment 2. The network segment 2 classifies the prediction sequence to obtain different fault types and transmits the fault types to the result output part. The result output part is used for training and diagnosis, and finally outputs a diagnosis result. The pyramid type time convolution neural network structure and the diagnosis method can fuse different sensor data, have small network parameters, require less memory resources during network training, and can perform real-time fault diagnosis.

The invention relates to a road slope rockfall detection method based on improved FCOS, and the method specifically comprises the following steps: 1, improving an FCOS algorithm, introducing a lightweight backbone network to replace an original feature extraction network lightweight network model, reducing network parameters, and training the network model to obtain a network model weight file; 2, quantifying the model into an edge end model of a calculation precision int8 type by using a PTQ quantification method, and deploying the edge end model to a horizon RDK X3 Module module; according to the method, a lightweight backbone network is used as a feature extraction network, network parameters are reduced, a model can be deployed on edge equipment, the model is quantitatively deployed to a horizon RDK X3 Module module for operation, accurate and real-time detection of road slope rockfall is realized, a data frame of an identified rockfall result is uploaded to a server, and the data frame of the rockfall result is analyzed. And the ground quality of the network and the transmission quantity of invalid data are reduced, and the pressure of the server is reduced.

This invention relates to noise reduction methods, and more particularly to a lightweight multi-axis Transformer single-channel noise reduction method. It can fully extract the latent time-frequency features of speech signals. A multi-head dynamic local self-attention module is employed to efficiently extract local features. The proposed method has fewer network parameters and lower computational cost, while remaining competitive with state-of-the-art methods in terms of speech signal quality and intelligibility. It can fully extract the latent time-frequency features of speech signals, effectively reducing computer resource consumption. Hint blocks are used to help the model better learn frequency feature information.

The present invention discloses a bionic target contour extraction algorithm based on deep learning. By combining the outer region suppression mechanism with the deep CNN framework, an edge detection model inspired by biological characteristics is established, and the edge detection model is used to extract the edge of the target image; the edge detection model includes a backbone network, a side output layer and an outer region suppression module SSM. The present invention achieves a better balance between the accuracy and efficiency of edge detection by designing an efficient backbone network, and designs an outer region suppression module SSM with a multi-layer convolutional layer stacking structure to increase the receptive field and characterize the spatial structure at different scales, effectively simulating the outer region suppression effect, and the SSM modulation operation can enable the network to integrate important information over a larger field of view. Moreover, the model of the present invention does not require pre-trained weights to assist in training, the network parameters are significantly reduced, and the reasoning speed under the same hardware conditions is faster. Compared with the prior art, the present invention is more efficient and lightweight.

The invention relates to the technical field of environment monitoring, and discloses a hazardous waste landfill leakage source detection method based on instance segmentation, comprising the following steps: S1, using an inspection robot to perform video acquisition at the bottom of a landfill, the video acquisition equipment comprising at least two visible light cameras, each camera being equipped with an infrared light supplementing device, and the camera being equipped with a camera light source; a dark light environment is adapted; s2, collecting forward and backward pictures through the camera; s3, performing instance segmentation processing on the acquired video by using a YOLOSeg-EViT network to obtain a mask mark and position information of a leakage source; and S4, generating a detection report containing the leakage source detection image and the corresponding coordinate information thereof. And landfill leakage source detection is carried out by adopting a lightweight instance segmentation network YOLOSeg-EViT. Compared with a target detection method, the instance segmentation method can generate the mask of the leakage source, so that the geometric features of the leakage source can be better known, and appropriate solutions can be taken.

The present application relates to the technical field of algorithm identification, in particular to a sheep group identity recognition method based on facial features, comprising constructing a sheep face original data set after de-redundancy and data enhancement, realizing sheep face area and three facial key point detection by using an improved RetinaFace network, extracting a 128-dimensional normalized sheep face feature vector through a HTL-Net network fused with TripletAttention, constructing a feature database, developing an intelligent platform integrating detection and recognition, and completing identity matching based on Euclidean distance. The method solves the problems of traditional ear tags, RFID and other recognition methods, such as easy to fall off, high cost, harm to sheep, low precision of existing recognition models, and difficult deployment, and has a sheep face detection accuracy of 96.15%, a recognition accuracy of 91.90%, a lightweight model, supports multi-form real-time recognition, and is suitable for large-scale sheep group precision management requirements.

The invention provides a machine vision image processing smoothing method and system. The method comprises the following steps: acquiring an original machine vision image, and performing preliminary noise reduction processing on the original machine vision image through a fast non-local mean algorithm and a box filter to acquire an image after preliminary noise reduction; performing end-to-end low-resolution image super-resolution reconstruction on the image subjected to preliminary noise reduction through a deep residual convolutional network to obtain a super-resolution reconstructed image; carrying out image edge analysis screening on the super-resolution reconstructed image to screen out an edge fuzzy gray level out-of-tolerance image; and carrying out edge blur elimination and gray level recovery on the edge blur gray level out-of-tolerance image by selecting an edge smoothing preserving model.

The present invention discloses a method, device, readable storage medium and equipment for silent liveness detection of human face, which belongs to the field of human face recognition. The method obtains a human face image to be detected and performs preprocessing; the preprocessed human face image is input into a pre-trained multi-cascade network to determine whether the human face image is alive; wherein the multi-cascade network includes a multi-level neural network cascaded front and back, each level of neural network is used for liveness judgment of a prosthesis attack type, when the current level neural network judges it as liveness, it enters the next level neural network, when any level neural network judges it as a prosthesis, the human face image is judged as a prosthesis, and when the last level neural network judges it as liveness, the human face image is judged as liveness. The present invention adopts a deep learning method and uses multiple shallow network cascades to perform silent liveness detection of human face, which improves the accuracy and efficiency of liveness detection and greatly improves the stability of the algorithm.

An image segmentation method based on an improved U2-Net network includes the following steps: 1) preprocessing the image; 2) performing adaptive histogram smoothing on the image; 3) data augmentation; 4) obtaining output 1; 5) obtaining output 2; 6) obtaining output 3; 7) obtaining output 4; 8) obtaining output 5; 9) obtaining output 6, a significant probability map 6; 10) obtaining output 7, a significant probability map 5; 11) obtaining output 8, a significant probability map 4; 12) obtaining output 9, a significant probability map 3; 13) obtaining output 10, a significant probability map 2; 14) obtaining a significant probability map 1; 15) mapping significant probability maps 1 to 6 to pixel values ​​between 0 and 1 after passing them through a convolutional layer and a sigmoid activation function, representing the probability that each pixel belongs to the foreground (i.e., an object), thus obtaining the segmentation map. This method reduces the manpower and material resources required for manual processing, reduces computational load, and improves segmentation accuracy.