Abstract: A cross-media search method using a VGG convolutional neural network (VGG net) to extract image features. The 4096-dimensional feature of a seventh fully-connected layer (fc7) in the VGG net, after processing by a ReLU activation function, serves as image features. A Fisher Vector based on Word2vec is utilized to extract text features. Semantic matching is performed on heterogeneous images and the text features by means of logistic regression. A correlation between the two heterogeneous features, which are images and text, is found by means of semantic matching based on logistic regression, and thus cross-media search is achieved. The feature extraction method can effectively indicate deep semantics of image and text, improve cross-media search accuracy, and thus greatly improve the cross-media search effect.

Abstract: Embodiments of the present disclosure provide an ion mobility spectrometer device. The ion mobility spectrometer device includes: an ion mobility tube, a sampling device, and a sampling and circulating gas path. The sampling device includes a solid sample desorption device and a gas sampling device. The solid sample desorption device is configured to process the solid sample into a first mixed gas containing the solid sample, and the gas sampling device is configured to process the gas sample into a second mixed gas containing the gas sample. The sampling and circulating gas path is configured to transfer the first mixed gas and/or the second mixed gas into the ion mobility tube for detection.

Abstract: A method for image matching includes acquiring a template image and a target image; acquiring a group of template features according to the template image; extracting a group of target features according to the target image; and according to template features and target features, calculating an degree of image similarity between the template image and each frame of target images, and using a target image with the maximum degree of image similarity as a matched image to the template image. In the image-matching method, image matching is performed by calculating an degree of image similarity between a template image and each target image according to a degree of image similarity between template features and target features, so that non-redundancy of features in an image matching process and correct image matching can be guaranteed, and the image matching accuracy can be improved.

Abstract: A microneedle chip and manufacturing method. The method comprises injecting, into a female mold, a fluid needle liquid, wherein forming cavities matching the shapes of needles of a microneedle chip are provided at the female mold and form a cavity array, injection inlets are provided at a surface of one side of the female mold, and air ejection openings are provided at a surface of another side of the female mold to form an air ejection surface; covering the air ejection surface of the female mold using a breathable film, and during injection, passing a gas through the breathable film so as to retain the liquid inside the forming cavities; curing the fluid needle liquid to form the microneedle chip, and demolding to obtain the same. By employing the air ejection openings and the breathable film, a liquid is retained while ejecting a gas, providing a favorable micro-injection effect.

Abstract: Provided is a method for detecting a violent incident in a video based on a hypergraph transition model, comprising a procedure of extracting a foreground target track, a procedure of establishing a hypergraph and a similarity measure, and a procedure of constructing a hypergraph transition descriptor; using the hypergraph to describe a spatial relationship of feature points, in order to reflect attitude information about a movement; and modelling the transition of correlative hypergraphs in a time sequence and extracting a feature descriptor HVC, wherein same can effectively reflect the intensity and stability of the movement. The method firstly analyses the spatial relationship of the feature points and a transition condition of a feature point group, and then performs conjoint analysis on same. The method of the present invention is sensitive to disorderly and irregular behaviours in a video, wherein same is applicable to the detection of violent incidents.

Abstract: The present disclosure provides a device for collecting semi-volatile or non-volatile substance, including an air nozzle, a front cavity and a collecting body. The air nozzle is configured to eject air to a sample attachment surface. The front cavity has an upper port. The collecting body is sealingly connected to a lower end of the front cavity, inside of which is provided with a cylindrical cavity and a conical cavity arranged vertically coaxially, and bottom of which is provided with a sample outlet. The collecting body is provided with an air intake passage which is non-coplanar with respect to an axis of the cylindrical cavity and is disposed obliquely downward and inward. The collecting body is further provided with an air exhaust passage one end of which is a discharge port connected to the interior of the cylindrical cavity, the other end is connected to an air pump.

Abstract: Disclosed is a method for detecting pedestrians in an image by using Gaussian penalty. Initial pedestrian boundary box is screened using a Gaussian penalty, to improve the pedestrian detection performance, especially sheltered pedestrians in an image. The method includes acquiring a training data set, a test data set and pedestrian labels of a pedestrian detection image; using the training data set for training to obtain a detection model by using a pedestrian detection method, and acquiring initial pedestrian boundary box and confidence degrees and coordinates thereof; performing Gaussian penalty on the confidence degrees of the pedestrian boundary box, to obtain confidence degree of the pedestrian boundary box after the penalty; and obtaining final pedestrian boundary boxes by screening the pedestrian boundary boxes. Thus, repeated boundary boxes of a single pedestrian are removed while reserving boundary boxes of sheltered pedestrians, thereby realizing the detection of the pedestrians in an image.

Abstract: Disclosed is a hierarchical division-based point cloud attribute compression method. For point cloud attribute information, a new hierarchical division based coding scheme is proposed, wherein a frame of point cloud is adaptively divided into a “stripe-macroblock-block” hierarchical structure according to the spatial position and color distribution of the point cloud, and stripes are coded independently from one another, increasing the coding efficiency, enhancing the fault tolerance of a system and improving the performance of point cloud attribute compression. The method comprises: (1) inputting a point cloud; (2) division of a k-dimension (KD) tree structure of the point cloud; (3) continuity analysis of point cloud attribute information; (4) stripe division of the point cloud; (5) division of macroblocks and coding blocks of the point cloud; and (6) intra-frame prediction, transformation, quantification and entropy coding based on a block structure.

Abstract: Disclosed is a multi-angle adaptive intra-frame prediction-based point cloud attribute compression method. A novel block structure-based intra-frame prediction scheme is provided for point cloud attribute information, where six prediction modes are provided to reduce information redundancy among different coding blocks and improve the point cloud attribute compression performance. The method comprises: (1) inputting a point cloud; (2) performing point cloud attribute color space conversion; (3) dividing a point cloud by using a K-dimensional (KD) tree to obtain coding blocks; (4) performing block structure-based multi-angle adaptive intra-frame prediction; (5) performing intra-frame prediction mode decision; and (6) performing conversion, uniform quantization, and entropy encoding.

Abstract: Provided is a method for detecting a violent incident in a video based on a hypergraph transition model, comprising a procedure of extracting a foreground target track, a procedure of establishing a hypergraph and a similarity measure, and a procedure of constructing a hypergraph transition descriptor; using the hypergraph to describe a spatial relationship of feature points, in order to reflect attitude information about a movement; and modelling the transition of correlative hypergraphs in a time sequence and extracting a feature descriptor HVC, wherein same can effectively reflect the intensity and stability of the movement. The method firstly analyses the spatial relationship of the feature points and a transition condition of a feature point group, and then performs conjoint analysis on same. The method of the present invention is sensitive to disorderly and irregular behaviours in a video, wherein same is applicable to the detection of violent incidents.

Abstract: A method and apparatus for inputting data for an electronic data entry device are provided. In one embodiment, identification of an input object such as the particular fingers of a user that are used to actuate a key region is performed. The symbol associated with the actuated key region and the finger (or other input object) used is determined. In other embodiments, virtual input devices with interfaces such as QWERTY style keyboards, phone keypads, and multi-touch capable touchpads or tablets are provided in input regions. One or more video capturing devices remotely acquire actuation information from the input regions during data entry. User inputted symbols or functions are determined based on the actuations, their locations and identified input object sets that caused the actuations.

Abstract: A sampling tube bundle includes a tube bundle body and a joint assembly. The tube bundle body includes a gas sampling tube for transmitting sampling gas; a metal woven layer wrapping an outer circumference of the gas sampling tube; a heating pad comprising a heat tracing cable wrapping an outer circumference of the metal woven layer, for heating the sampling gas in the gas sampling tube; a flame resistant tape wound around an outer circumference of the of the heating pad; a plurality of signal transmission lines circumferentially spaced apart from one another are arranged outside the flame resistant tape; and a sheath located outside the signal transmission line. The joint assembly is arranged on an end of the tube bundle body, for connecting to an analysis device or a sampling device.

Abstract: The present invention discloses a method for enhancing a low-illumination image based on a camera response characteristic, including: selecting an exposure model and obtaining a camera response equation model corresponding thereto; determining a parameter of the camera response equation model; estimating an exposure ratio of a multi-exposure image sequence to be generated to an original image so as to generate the multi-exposure image sequence; and performing image fusion on multi-exposure images to obtain an enhanced image with a better visual effect and less distortion. The method can solve the problem that existing low-illumination image enhancement algorithms introduce more artificial traces while enhancing images, and can obtain the enhanced image with the better visual effect and less distortion, thereby obtaining the enhanced image preserving naturalness.

Abstract: Disclosed are a shadow removing method for an image and an application. The shadow removing method comprises a shadow-free feature analysis process, a shadow-free transformation parameter acquisition process and a shadow-free feature imaging process. The application is for road surface detection, a detection method comprising: firstly, using a shadow-free feature extraction method to select a region of interest and extract a feature; next, performing image filtering, segmentation and road surface region selection; lastly, performing image morphology filtering and hole filling. The method may remove shadows in color images, thus serving as a pre-processing step applied in various machine vision fields, and the application in road surface detection may solve the problem of detecting a road surface in dark shadows. The method has the advantages of having low complexity, high processing speed and high accuracy.

Abstract: A Collaborative Deep Network model method for pedestrian detection includes constructing a new collaborative multi-model learning framework to complete a classification process during pedestrian detection; and using an artificial neuron network to integrate judgment results of sub-classifiers in a collaborative model, and training the network by means of the method for machine learning, so that information fed back by sub-classifiers can be more effectively synthesized. A re-sampling method based on a K-means clustering algorithm can enhance the classification effect of each classifier in the collaborative model, and thus improves the overall classification effect.

Abstract: A video action detection method based on a convolutional neural network (CNN) is disclosed in the field of computer vision recognition technologies. A temporal-spatial pyramid pooling layer is added to a network structure, which eliminates limitations on input by a network, speeds up training and detection, and improves performance of video action classification and time location. The disclosed convolutional neural network includes a convolutional layer, a common pooling layer, a temporal-spatial pyramid pooling layer and a full connection layer. The outputs of the convolutional neural network include a category classification output layer and a time localization calculation result output layer. The disclosed method does not require down-sampling to obtain video clips of different durations, but instead utilizes direct input of the whole video at once, improving efficiency.

Abstract: Provided is a point cloud attribute compression method based on a KD tree and optimized graph transformation, wherein same, with regard to point cloud data, reduces the influence of a sub-graph issue on the graph transformation efficiency by means of a new transformation block division method, optimizes a graph transformation kernel parameter, and improves the compression performance of the graph transformation, and comprises: point cloud pre-processing, point cloud KD tree division, graph construction in the transformation block, graph transformation kernel parameter training, and a point cloud attribute compression process.

Abstract: The present disclosure relates to the technical field of safety detection, and in particular to a sample collecting and introducing device and a detection system. The sample collecting and introducing device provided by the present disclosure includes a sampling device for collecting a sample, and a semipermeable membrane device for extracting the sample collected by the sampling device and conveying the extracted sample to detection equipment, wherein the sampling device is provided with an air guide cavity, the air guide cavity is configured to guide airflow carrying the sample to flow to the semipermeable membrane device, the semipermeable membrane device is provided with a semipermeable membrane which is arranged outside the sampling device. In the present disclosure, the size of the semipermeable membrane is no longer limited by the sampling device, and therefore the difficulty of increasing the area of the semipermeable membrane is reduced.

Abstract: A method and apparatus for inputting data for an electronic data entry device are provided. In one embodiment, identification of an input object such as the particular fingers of a user that are used to actuate a key region is performed. The symbol associated with the actuated key region and the finger (or other input object) used is determined. In other embodiments, virtual input devices with interfaces such as QWERTY style keyboards, phone keypads, and multi-touch capable touchpads or tablets are provided in input regions. One or more video capturing devices remotely acquire actuation information from the input regions during data entry. User inputted symbols or functions are determined based on the actuations, their locations and identified input object sets that caused the actuations.

Abstract: Provided is a point cloud attribute compression method based on a KD tree and optimized graph transformation, wherein same, with regard to point cloud data, reduces the influence of a sub-graph issue on the graph transformation efficiency by means of a new transformation block division method, optimizes a graph transformation kernel parameter, and improves the compression performance of the graph transformation, and comprises: point cloud pre-processing, point cloud KD tree division, graph construction in the transformation block, graph transformation kernel parameter training, and a point cloud attribute compression process.