35results about How to "Fully consider the relevance" patented technology

An electroencephalogram (EEG) feature extraction method based on dominant electrode combination and empirical mode decomposition (EMD) includes inputting n-channel EEG signal data, selecting dominant electrodes according to the principle that electrodes with record of EGG signals having classification performance higher than a certain threshold are called the dominant electrodes, otherwise, are called non-dominant electrodes, selecting dominant combinations, respectively extracting feature of EGG data of a training sample set and EGG data of a testing sample set corresponding to each dominant combination by means of the EMD to acquire training feature vector and testing feature vector of each dominant combination, respectively inputting the training feature vector, the testing feature vector, training sample set label, testing sample set label of each dominant combination input a naive Bayesian classifier to classify to obtain classification accuracy of each dominant combination, and finally surmising relation between different stimulations and activated brain regions when executing relevant motor imagery tasks according to the classification accuracy of each dominant combination.

The invention discloses an energy spectrum CT image quality evaluation method. The method includes the steps of obtaining 101 mono-energy images in an energy range of 40 to 140keV (kiloelectron volt) through energy spectrum CT images; determining an interesting organization structure and a background structure in every mono-energy image and respectively selecting interesting regions ROI of the same area in the interesting organization structure and the background structure; respectively calculating the mean value <t>1 and the variance sigma (2)1 of the interesting region ROI of the interesting organization structure, the mean value <t>0 and the variance sigma (2) 0 of the interesting region ROI of the background structure, and the pseudo ideal observer signal to noise ratio (PIO SNR) of the interesting organization structure of every mono-energy image; drawing a changing curve of the PIO SNR changing with the energy according to the pseudo ideal observer signal to noise ratio (PIO SNR) of every mono-energy image and determining a optimum energy point. The energy spectrum CT image quality evaluation method can quickly determine the optimum energy point of the interesting region, takes the relevance between the background noise and the noise of the interesting region into full consideration, is close to and in line with characteristics of a human visual system, and has the advantages of stability and efficiency.

The invention provides a culture medium for germinating boletus spores, and the culture medium comprises a dried chroogomphus rutilus product, a PDA (Potato Dextrose Agar) culture medium and water. The invention also provides a method for preparing the culture medium and a method for germinating the boletus spores by adopting the culture medium. The culture medium provided by the invention has the advantages of simple formula, easiness for preparation, more spore germination, regular colonial morphology, thick and dense hypha, and the like.

The invention involves a screening method of a statistical reduction standard model forecasting factor, including the following steps: step 1, establish a location relationship diagram of the watershed meteorological site and grid; step 2, push the forecast volume and re -analysis of the predictions such as precipitation, temperature and temperature of each site.The correlation between data atmospheric circulation factor, including the correlation with the grid where the meteorological site is located, and the correlation with the surrounding grid, select suitable factor according to the set correlation threshold;At the three standards of seasonal and annual standards, the regular determination standard determines the change of sexual coefficient; step 4, analyze and test the model according to the different scale and different selection factor methods of step 3.

The invention discloses an auxiliary vehicle traveling path system based on a mobile terminal. The auxiliary vehicle traveling path system comprises a vehicle flow detecting module, area servers, a top layer server, an application server and an application client side. A camera module is directly connected with an image detecting module to form the flow detecting module. A plurality of detecting modules in one area are connected with the area servers through a router, and the area servers can be arranged in multiple layers according to the size of the area. All the area servers in different areas are connected with one top layer server through the router. All top layer servers under one subnet are connected with the application server through the router. The application server is connected with the internet through the router. The application client side interacts with the application server through the internet. The flow of vehicles in a large range can be monitored in the manner that a layering type networking mode is adopted in the vehicle flow real-time monitoring network, a traffic regulation department is helped to carry out traffic dispatching, and people can be helped to select the optimal driving path.

The invention discloses an underwater video compression method, which comprises intraframe compression and interframe compression of an image to be compressed. The image to be compressed is YUV spatial data. The method also comprises the step of wavelet transformation pretreatment of the image to be compressed before intraframe compression and interframe compression. The wavelet transformation pretreatment comprises the following steps: conducting at least first-order integer wavelet transformation of the Y-component of the image to be compressed, and rejecting at least one high-frequency component selected from LH, HL and HH. The method rejects at least part of the high-frequency components of the Y-component after wavelet transformation of the image to be compressed and remains the low-frequency component of the U and V components after wavelet transformation, so as to reduce data quantity to be encoded, thereby increasing compression ratio from another aspect under the premise of ensuring the image quality.

The present invention discloses a recommendation method and apparatus based on the heterogeneous information network. The method is characterized in that: a schematic diagram of the heterogeneous information network is divided into a plurality of sub-diagrams, each of the sub-diagrams corresponds to a unary diagram; based on each of a plurality of user-item similarity matrices, fusion is carried out on each of the user-item similarity matrices, and a user-item scoring equation is constructed; and by optimizing the objective function, the user-item scoring equation is trained according to the known scoring sample to obtain user-item scoring equation parameters, so that scoring of the user for each item can be predicted, the item whose scoring meets the preset condition is recommended to theuser, the problem that the semantic coverage is incomplete when using the element path to analyze similarity between users and items in the prior art is solved, the relevance between features is fully considered, and useful products are recommended to the users.