41results about How to "Improve the evaluation index" patented technology

The invention provides a natural scene horizontal character detection method based on a deep convolutional neural network. According to the method, deep optimization is carried out on the basis of a TextBoxes network model, a new text prediction convolution group is added, and the network depth is expanded, so that the feature learning of the network for a small data set is more sufficient, and the feature information of a plurality of convolution layers is fully utilized to carry out fusion learning under certain model complexity. After feature learning is performed on original picture data through convolution layers with different receptive fields, a text prediction layer is utilized to return to the position of a textbox and predict a text category. According to the detection method, the influence of factors such as background complexity of a natural scene and insufficient features of a small data set on character detection is effectively solved. Experimental verification is carriedout under a Caffe platform, and results show that the model can effectively improve the recall rate and comprehensive evaluation indexes of natural scene level character detection under a small dataset.

The invention relates to a visual attention detection method for a three-dimensional video. The method is characterized in that firstly a characteristic contrast is calculated by utilizing a low-level visual characteristic so as to obtain spatial saliency of a three-dimensional video frame; and in addition, time saliency is obtained by utilizing motion information, and motion saliency is calculated by using a motion in plane and a motion in depth for the motion information in the three-dimensional video. Finally a saliency map is obtained by the spatial saliency and the time saliency, and a common fate law and a tightness law in Getalf psychology are applied in a combination process. An experimental result shows that a good effect is achieved in saliency prediction of the three-dimensional video.

The invention provides a cross-social-network user identity recognition method based on a neural tensor network. The method comprises the following steps: step 1, based on network representation learning of Random Walks and Skipgram models, mapping network structure spaces of a source network Gs and a target network Gt to vector spaces respectively; step 2, based on the vector space obtained in the step 1, modeling an association relationship between the user nodes in the source network Gs and the target network Gt by using a neural tensor network model; and 3, inputting the incidence relationvector obtained by modeling in the step 2 into a multi-layer perceptron model for dichotomy, and judging whether the user node pairs between the source network Gs and the target network Gt point to the same real user or not according to a classification result. According to the method, the neural tensor network model is adopted to replace a standard neural network model, the model has stronger capability of expressing the relationship among cross-network users, and two user vectors can be associated in multiple dimensions.

The invention discloses an answer ordering method for a community question-answer platform. According to the method, rich metadata, such as themes, timestamps and text content of questions and answers, of the questions and answers can be sufficiently used for solving the answer ordering problem; meanwhile, an enhancement-type attention mechanism recurrent neural network model (EARNN) is used for solving the answer ordering problem, and compared with a tradition model, more information is used. A predicted result is improved to a certain degree on multiple evaluation indexes.

The invention relates to a two-dimensional video saliency detection method based on long-term and short-term memory. The method is characterized by comprising the following steps: firstly, extractingshort-term time sequence features by utilizing a 3D convolutional network (3D-ConvNet); secondly, extracting long-term time sequence features by adopting a bidirectional long-term and short-term memory network (B-ConvLSTM); fusing the extracted short-term time sequence features and long-term time sequence features; and finally, obtaining a saliency map through fusion result deconvolution. The model is combined with long-term and short-term time sequence characteristics, the operation information of the salient target in the video can be effectively reserved, and the saliency prediction experiment result of the two-dimensional video proves that the proposed model can obtain a good detection effect.

The invention discloses an integrated learning-based software defect reopening prediction method. The method comprises the following steps of: S1, extracting an LWEparagraph2vec-based semantic vector features from a defect report of software; S2, combining the LWEparagraph2vec-based semantic vector features extracted from the defect report of the software with meta features thereof to form a feature set; S3, constructing a prediction model according to an imbalanced data processing-based integrated learning prediction algorithm UnderSMOTEBagging method; and S4, obtaining a class label of a living example according to the feature set extracted in the step S2 and the prediction model obtained in the step S2, so as to judge whether detects of the software is going to be reopened or not. The method disclosed by the invention is capable of solving the problem that the prediction effect is not ideal due to data imbalance in software defect reopening prediction and finiteness of the used feature set.

The invention provides a machine reading understanding model training method and device based on a joint loss function. Specifically, when machine reading understanding model training is carried out,a loss function composed of a maximum likelihood estimation function and a minimum risk training function is used as an evaluation index of a machine reading understanding model so as to guide adjustment of parameters of the machine reading understanding model. The idea of the minimum risk training function is to use the loss function to describe the difference degree between the answer output bythe model and the standard answer. The maximum likelihood estimation function, namely loss, tries to find a group of model parameters, so that the loss value of the machine reading understanding modelon the training set is minimum, and therefore, compared with the mode of singly utilizing the maximum likelihood estimation function, the model trained by the method provided by the invention can extract answers more accurately.

The invention discloses a synthetic aperture radar (SAR) image compression method based on target area extraction and direction wave, and mainly solves the problems of less edge information stream distribution and loss of important information due to the fact that the same transform compression strategy is used in an image target area and a background area in the existing method. The method comprises the implementation steps of extracting a texture map of an SAR image by a variation coefficient; carrying out quadtree partitioning on the SAR image, dividing an image block into the target area and the background area by the texture map; detecting a diverting pair of the target area by the texture map, and carrying out pruning processing on partitioned images; and carrying out directionlets transformation on the target area, carrying out wavelet transformation on the background area, and respectively encoding the coefficients of the target area and the background area by a set partitioning in hierarchical trees (SPIHT) encoding method. The SAR image compression method has the advantage of well protecting the information of the target area by different transform compression strategies in the target area and the background area, and can be applied to real-time transmission and storage of the SAR image.

The invention discloses a method for predicting a next track point of a user. The method comprises the following steps: crawling a certain amount of data: an ID of the user, position information of aseries of short-term and long-term historical track points corresponding to the user, and a timestamp of each track point; constructing a feature interaction self-attention network model based on thecrawled information, and making attention in combination with a result that the position information of the long-term historical track points of each user passes through a self-attention layer; performing optimal training on the parameters by using a cross entropy loss function; for a new user and a series of historical track points thereof, and constructing a series of instances by utilizing theID information, the position information of a series of historical track points corresponding to the user and the timestamp of each track point, and inputting the instances into a trained feature interaction self-attention network model, thereby obtaining a series of sorting scores of predicted positions. According to the method, the problem of predicting the next track point by utilizing the richmetadata of the user and the historical track is solved, and the prediction accuracy is greatly improved.

The invention discloses an image super-resolution reconstruction method. The method includes two parts: 1. Adaptive decomposition of the image. The so-called self-adaptation here refers to the self-adaptive drive decomposition process of the image data. 2. Perform super-resolution amplification on the basis of image adaptive decomposition. On the basis of adaptively decomposing the image into low-frequency information and high-frequency information, a Gaussian mixture model is used as a priori model in the high-frequency part for super-resolution amplification, and spline interpolation is used in the low-frequency part, and finally superimposed. The invention can realize high-quality enlargement of images.

Belonging to the technical field of drugs, the invention discloses a release detection method of an aniracetam sustained release tablet. The release detection method of the aniracetam sustained release tablet comprises the steps of: 1) in vitro dissolution rate test of the aniracetam sustained release tablet; 2) determination of in vivo blood concentration; and 3) in vitro-in vivo correlation analysis. The invention constructs in vitro evaluation model simulating the in vivo, establishes an in vitro-in vivo correlation multi-criteria evaluation technology, increases new evaluation indexes, and establishes a new detection method, better explains advantages of the prescription, establishes an evaluation technique suitable for drugs with too quick prototype drug release, and provides the basis and guidance for study of the drug in future.

The invention discloses a sequence position recommendation method based on geographic perception. The problem of position recommendation can be solved by fully utilizing rich metadata (user ID, timestamp, position and place longitude and latitude information) of users and historical behavior tracks. Meanwhile, the geographic perception sequence recommender GeoSAN based on the self-attention network is used for recommending the user position, compared with a traditional model, geographic information is more fully utilized, the data sparsity problem can be well solved, and the recommendation result is improved to a certain extent on multiple evaluation indexes.

The invention relates to the technical field of machine learning, in particular to a negative sample extraction method and device, computer equipment and a storage medium, and the method comprises thesteps: obtaining the page burying point information of a display page in an application platform, and determining the label information and popularity information of each burying point object according to the page burying point information; acquiring historical behavior information of a user in the application platform, and determining a label weight of the user in the application platform according to the historical behavior information; according to the label weight and the label information of each buried point object, determining the sampling probability of each buried point object sampled by the user in the application platform; generating a negative sample distribution sequence according to the popularity information and the sampling probability of each buried point object, and extracting a negative sample from the negative sample distribution sequence. According to the scheme, the sampling logic of the negative sample is optimized, the calculation amount in the model training process is reduced, and the model effect and the evaluation index are improved.

The invention discloses an air-ground cooperative vehicle positioning and orienting method, and particularly relates to the technical field of vehicle positioning and orienting method application. An air-ground cooperative vehicle positioning and orienting method comprises the following steps that an unmanned aerial vehicle is adopted to mount industrial cameras to construct a non-overlapping view field camera system, coordinate transformation between the cameras is solved, and internal parameters and external parameters of the multiple cameras are obtained; internal parameters and external parameters obtained in the camera system are classified and recognized based on marker detection of a structural forest and PCANet, so that position information of markers is obtained; and S2, the orientation of the vehicle is estimated according to a visual reckoning positioning algorithm of time-space consistency in combination with the position information of the marker detected in S2, and the position and state information of the vehicle in real time is estimated by adopting state filtering. By the adoption of the technical scheme, the safety problem of large engineering vehicles and special vehicles during complex road operation is solved, and effective vehicle position information can be provided for drivers.

The invention provides an abstract method based on a social media microblog specific topic. The method comprises the following steps of (1) obtaining W according to a following formula in the description, wherein the formula merges a microblog abstract optimization model of group sparsity study and social regular term parameters; the S is a text matrix; a data set is obtained through the calculation from TF-IDF; the W is a refactoring coefficient matrix; Lambda is a group sparsity regular term parameter; the L is D-T and is an Laplace matrix; (2) obtaining the importance Score (i) of the i-thmicroblog by calculating the normal formulas of the i-th line of the W shown in the description; sequencing the microblog according to the importance; further screening the front k microblogs to be used as the abstract, wherein the Score (i) is shown in the description. The abstract method is based on the basic framework of the sparse reconstruction and merges the social media content and a socialnetwork structure; the obtained microblog abstract is more similar to an expert mutual evaluation result in aspects of three evaluation indexes of ROUGE-1, ROUGE-2 and ROUGE-SU4 through being compared with the existing model.

The invention relates to a multi-mode video dense event description algorithm of an interactive Transform, and belongs to the technical field of video algorithms. The method comprises the following steps: 1, extracting visual features, audio features and voice features in a video; information in the video is better utilized through multi-modal feature extraction; 2, fusing the visual features with the audio features and the voice features through an interactive attention module in the interactive Transform, and further coding the video features; 3, completing model training in two stages; firstly, a description model is trained based on real video segments, then the encoder weight of the trained description model is frozen, and then a segment proposal model is trained. According to the method, the feature information in the video is fully utilized, the multi-modal features are interactively fused, and a good dense video description effect is shown.

The invention discloses an underwater image enhancement method fusing deep learning and a traditional image enhancement technology, which comprises the following steps of: firstly, analyzing an average value difference between each channel of an input underwater image and a corresponding natural image, and showing that a red channel needs to be compensated and a green channel needs to be attenuated from the difference; therefore, the color compensation is performed on the R channel and the G channel of the input underwater image by using an attention-guided residual network, and the motivation of the strategy is that most of the underwater images are observed to be composed of relatively single and uniform color distribution. For scene contrast enhancement and scene deblurring, a multi-scale convolutional neural network is developed, and a CLAHE (Contrast Adaptive Histogram Equalization) and a Gamma correction algorithm are introduced as supplements to process a complex and changeable underwater imaging environment. Experimental results show that the underwater image enhancement method has a good effect.

The invention provides a remote sensing image text generation and optimization method based on self-reinforcement learning. The remote sensing image text generation and optimization method based on self-reinforcement learning comprises the following steps: S1, extracting remote sensing image semantic understanding feature; S2, obtaining a training set, pre-training the text generation model, and extracting text generation model parameters; and S3, inputting the extracted feature vectors, a prior text library, pre-trained text generation model parameters and task requirements of a user into a remote sensing image text generation network, and restoring image feature information represented by the extracted feature vectors into text description through a deep learning natural language processing technology. The text is generated by adopting the self-reinforcement learning remote sensing image text generation algorithm based on the strategy gradient algorithm, the training effect of the remote sensing image generation model is improved, parameters are promoted to converge towards expected values, and the accuracy of generation description is improved.

The invention provides a single image rain removal method based on a multi-scale fusion generative adversarial network. The method comprises the following steps of firstly, detecting a rain image through image saliency to obtain a saliency map, fusing the saliency map with a rain image Concat, and accurately recognizing a raindrop area to be restored, then, using multi-scale fusion to generate anadversarial network for rain removal, conducting multi-scale fusion on l12, l14 and l16 layers of networks in a generator network, enhancing the quality of a generated rain removal image, enabling thediscriminator network to adopt combination of global discrimination and local discrimination, and conducting training to obtain a final network model, and inputting the test set into the trained model to obtain a rain removed image, and evaluating the generated image according to the SSIM and PSNR indexes. The image generated by the image rain removal method provided by the invention is better invisual effect, the removed raindrop area is more authentic and coherent, and each evaluation index is improved.

The invention relates to the field of function testing, and in particular relates to a testing method for a noise reduction function of a device and the relevant device. The testing method for the noise reduction function of the device comprises the following steps: acquiring performance parameters of a to-be-tested device, and establishing a testing scene of the to-be-tested device according to the performance parameters; acquiring parameters of the testing scene, according to the parameters of the testing scene, extracting any original sound corresponding to the parameters of the testing scene from a preset sound library, and according to the differences between a sound wave curve of the original sound and a preset sound wave curve, correcting the original sound, so that a sound sample is obtained; and inputting the sound sample into the to-be-tested device, carrying out playing through the to-be-tested device, receiving the played sound sample, thus obtaining a testing sample, comparing the testing sample with the sound sample, and generating a noise reduction testing report according to the comparing result. The evaluation index for testing the noise reduction function is increased, the testing dimensionality is expanded, the testing efficiency is promoted, and the automatic testing can easily access in quality testing.

The invention discloses a motion direction change-based candidate seed point cloud single target tracking method, which comprises a training module and a test module, and is characterized in that the training module firstly preprocesses point cloud to obtain template point cloud, then uses Gaussian sampling to obtain candidate point cloud, and then inputs the template point cloud and the candidate point cloud into an encoder for encoding, a corresponding feature vector is obtained, a distance loss function and a direction loss function are calculated, and the whole model is trained. The test module firstly uses a pre-trained PointRcnn model to carry out target detection, then carries out candidate area sampling, then inputs sampled candidate point cloud and tracking target point cloud of a previous frame into the trained model to carry out coding, and finally carries out target tracking on a coded feature vector by using cosine similarity comparison. According to the method, the single-target tracking precision can be improved, and a wrong tracking phenomenon can be effectively prevented.

The invention provides a retina vessel segmentation method based on residual network feature extraction, which is applied to a neural network model, and comprises the following steps: enabling an original retina vessel image to pass through a pre-trained VGG coding layer to obtain a plurality of images, wherein the number of the images is five, the sizes of the images have a preset proportional relation with the sizes of the original retinal blood vessel images. Therefore, the segmentation precision of the network is obviously improved, and the fitting ability and generalization ability of the model are better optimized. Compared with a Unet, the network also generally has better performance in other data sets.

The invention relates to a risk prediction method based on clinical examination and medication intervention data, and the method comprises the steps: selecting start and end nodes from the clinical examination data in an individual observation period, carrying out the vectorization modeling, and obtaining an input vector x1; constructing an intervention dictionary, calculating the characteristic frequency of medication intervention, and performing vectorization modeling on individual medication intervention data to obtain an input vector x2; combining the input vector x1 and the input vector x2 to obtain an input feature vector X; inputting the input feature vector X into a prediction model, obtaining a real result Y through fitting, optimizing prediction model parameters, and obtaining a final prediction model; inputting the individual data into the final prediction model subjected to parameter adjustment and outputting model prediction results. According to the method, the design is reasonable, the relationship between different medication intervention combinations and the influence of the medication intervention combinations on the individual state can be explored, the prediction is accurate and reliable, and each evaluation index is improved.

The invention discloses a social text-oriented cascade linear entity relationship extraction method, which adopts a cascade linear extraction mode, that is, the method firstly detects relationships contained in a given text, and then takes each relationship as additional knowledge to guide the subsequent extraction process of a head entity and a tail entity. According to the method, two decoders, namely a relation decoder and an entity decoder, are further designed, and the two decoders are jointly used for extracting the entity relation triad. By means of the method, the extraction accuracy of entity pairs (the head entity and the tail entity) and the performance of combined extraction can be improved, the overlapping problem can be naturally solved through a relation-first cascade extraction method, and then a more accurate premise can be provided for construction of the knowledge graph.

The invention provides a student score prediction method based on a two-way attention mechanism. The student score prediction method comprises the following steps of obtaining student attribute characteristics, first-stage historical scores and second-stage historical scores; obtaining a feature vector f1 of each student attribute feature with respect to a first-stage historical score and a feature vector f2 of each student attribute feature with respect to a second-stage historical score; fusing the feature vector f1 and the feature vector f2 to obtain a feature vector f; and calculating a student score prediction score p from the feature vector f through a three-layer full-connection neural network MLP. By introducing a double-path attention mechanism, the defects of a single-path attention mechanism can be overcome, information complementation is well performed, and the prediction performance of the model is further improved; and for a prediction result, a plurality of evaluation indexes are remarkably improved.

The invention discloses an ultra-sparse CS fusion method applied to structured light images. Step 1: When the images captured by a structured light device are multiple images with unclear details or missing details, the images are transmitted to a computer; Step 2: Decompose each image by wavelet; step 3: perform compressed sensing compression on the larger wavelet coefficients decomposed under the compressed sensing framework, and fuse the compressed ultra-sparse coefficients by maximizing the absolute value; the fusion base is selected using the coif4 wavelet base ; Step 4: Through the orthogonal matching pursuit method and iterative calculation, the CS image restoration after super-sparse coefficient fusion is realized; Step 5: The restored image is transmitted back to the institutional optical imager through electrical signals for continued mechanical operation. It realizes under-sampling under the condition of signal sparse or nearly sparse, and then restores the defective structured light image with high probability through the best fusion method based on compressed sensing.

The invention provides a progressive die equipment and method for machining a sheet metal component with a threaded hole on a lateral-face edge belt, belonging to the field of metal machining. The equipment comprises an upper die and a lower die; the upper die and the lower die are mutually matched; and a plurality of punching machining positions are arranged on the upper die and the lower die. The progressive die equipment is characterized by being provided with a screw tap position; a screw rod is arranged on the upper die at the screw tap position; a screw tap fixing device is arranged on the lower die; the screw tap fixing device comprises a rotary mechanism engaged with the screw rod; and the rotary mechanism is connected with a screw tap. The progressive die equipment and the method provided by the invention have the advantages of high production efficiency and high precision; the one-step formation in a progressive die punching process can be realized; and the process can be simplified, the number of the equipment is reduced, and the energy consumption is reduced.

The invention discloses a relation perception similar problem identification and evaluation method, system and device and a storage medium, according to a related scheme, a similar problem identification model based on a relation perception neural network is used for carrying out similar problem identification of problem pairs, and compared with a traditional model, semantic relation related information of multiple semantic matching is used. For a prediction result, a plurality of evaluation indexes are improved to a certain extent.

The invention relates to a visual attention detection method for a three-dimensional video. The method is characterized in that firstly a characteristic contrast is calculated by utilizing a low-level visual characteristic so as to obtain spatial saliency of a three-dimensional video frame; and in addition, time saliency is obtained by utilizing motion information, and motion saliency is calculated by using a motion in plane and a motion in depth for the motion information in the three-dimensional video. Finally a saliency map is obtained by the spatial saliency and the time saliency, and a common fate law and a tightness law in Getalf psychology are applied in a combination process. An experimental result shows that a good effect is achieved in saliency prediction of the three-dimensional video.