684results about How to "Easy to train" patented technology

A discriminative grammar framework utilizing a machine learning algorithm is employed to facilitate in learning scoring functions for parsing of unstructured information. The framework includes a discriminative context free grammar that is trained based on features of an example input. The flexibility of the framework allows information features and/or features output by arbitrary processes to be utilized as the example input as well. Myopic inside scoring is circumvented in the parsing process because contextual information is utilized to facilitate scoring function training.

A process for facilitating a client system defense training exercise implemented over a client-server architecture includes designated modules and hardware for protocol version identification message; registration; profiling; health reporting; vulnerability status messaging; storage; access and scoring. More particularly, the server identifies a rule-based vulnerability profile to the client and scores client responses in accordance with established scoring rules for various defensive and offensive asset training scenarios.

The invention provides an image denoising method based on generative adversarial networks, and belongs to the technical field of computer vision. The method comprises the following steps: (1) designing a neural network for estimation for noise intensity of an image containing noises; (2) using image blocks in an image library to add noises of the intensity according to the estimated noise intensity to use the same as samples of training the networks; (3) in network training, designing a new generation network and discrimination network, and adopting a form of fixing the generation network to train the discrimination network and fixing discrimination network parameters to train the generation network to enable the networks to carry out adversarial training; and (4) using the trained generation network as a denoising network, and selecting a network parameter according to a result, which is obtained by the noise recognition network, to denoise the image containing the noises. The methodhas the effects and the advantages that a visual effect of the denoised image is improved without the need for manual intervention for adjusting the parameter, and texture details of the image can bebetter restored.

The invention discloses an ReLU convolutional neutral network-based image denoising method. The method comprises the following steps of building an ReLU convolutional neutral network model, wherein the ReLU convolutional neutral network model comprises a plurality of convolutional layers and active layers after the convolutional layers, wherein the active layers are ReLU functions; selecting a training set, and setting training parameters of the ReLU convolutional neutral network model; training the ReLU convolutional neutral network model by taking a minimized loss function as a target according to the ReLU convolutional neutral network model and the training parameters of the ReLU convolutional neutral network model to form an image denoising neural network model; and inputting a to-be-processed image to the image denoising neural network model, and outputting a denoised image. According to the ReLU convolutional neutral network-based image denoising method disclosed by the invention, the learning ability of the neural network is greatly enhanced, accurate mapping from noisy images to clean images is established, and real-time denoising can be realized.

The invention provides a remote unmanned aerial vehicle cluster control method and system based on 3G (the 3rd Generation Telecommunication) and GPRS (General Packet Radio Service) cell phone communication and belongs to the technical field of unmanned aerial vehicles. Each unmanned aerial vehicle is provided with an independent unique electronic identification code by integrating a cell phone communication module into a flight control computer board card; as long as the system is started and electrified, the current relevant information of the unmanned aerial vehicle is transmitted to a remote control center; a permission instruction is necessarily applied from a control center for each flight; information such as an airway plan is made uniformly in the control center; and the information is transmitted to an operation terminal of the unmanned aerial vehicle through a cell phone communication module. Consequently the unmanned aerial vehicle cannot fly away without a headquarter control instruction. The state of each unmanned aerial vehicle is fed back to the control center and can be shared with networked data of relevant national supervision departments so as to play roles in unified supervision and scheduling. In addition, the on-site operation personnel do not need professional knowledge such as airway point setup; and the unmanned aerial vehicle can be operated by common people subjected to simple training, so that the problem of staff shortage in unmanned aerial vehicle control is solved.

A method of controlling the operation of an animal training device which is worn around the neck of an animal to apply at least one of an impulse wave and vibration to the animal. Upon pushing a first function switch under the condition that a security code number is set by a security code setting unit, the level of the impulse wave to be applied to the animal is set by a volume adjustment unit and a mode selector is set to any one of a vibration position, vibration/impulse wave position, impulse wave position, a receiver receives a radio signal containing such data and thus applies at least one of the vibration and impulse wave to the animal to train it. Further, the volume adjustment unit employs a variable resistor which can finitely adjust the impulse wave level. Therefore, only a vibration or a low-level impulse wave can be applied to an animal over-sensitive to an electrical stimulation to readily train it.

The invention discloses an image super-resolution method based on SAE and sparse representation, and belongs to the field of image processing. The image super-resolution method mainly comprises an off-line training stage and a test refactoring stage, wherein in the off-line training stage, image characteristics extracted by an SAE (Sparse Auto Encoding) model are subjected to dictionary training, and a dictionary pair reflecting corresponding relations of high-resolution images and low-resolution images is established; in the test refactoring stage, low-resolution images inputted by a user are subjected to super-resolution reconstruction by the obtained dictionaries and a sparse representation method. Through the application of the image super-resolution method, unsupervised learning training is performed on original image sampling data by using the SAE model, so that the defects that manually designed operator characteristic extraction is time-consuming and strenuous and the extracted characteristics are single are avoided, meanwhile, image characteristics represented by SAE compression are directly used for training of the high-low-resolution dictionary pair, the dictionary training is facilitated, lost detail components in the images can be estimated by the sparse representation method, and higher-quality high-resolution images can be restored from the low-resolution images conveniently.

This invention discloses one forest measurement method without human aviation cameral remote sensor, which adopts aviation device as remote platform without human and integrated digital codes camera, difference GPS and top platform, wherein, it uses digital camera technique to get high earth resolution rate without human remove image with large proportion rules fully in use of high flexible machine without special plane spot and capable of flying under clouds.

An integrated respiratory monitor and imaging device apparatus (10) is provided. The apparatus is useful for establishing pre-operative and intra-operative breath hold congruency in patients and for other interventional work. The apparatus (10) includes a respiratory monitor system (12) and an imaging device (14). The respiratory monitor system is adapted to engage a patient and generate a respiratory signal representative of a breath hold level of the patient during a breath hold. The imaging device (14) is adapted to scan the patient during the breath hold and generate a volumetric image data set of the patient. The respiratory sensor and imaging device are operatively connected to associate the respiratory signal representative of the breath hold level of the patient together with the volumetric image data set of the patient. A data storage device (64) is provided for storing a set of respiratory signals in association with a corresponding set of volumetric image data sets in the subject apparatus (10). A scanner gating function is provided to center the image acquisition time at selected points in the respiratory cycle such as at a minimum to minimize motion artifacts in the resultant image.

A method is provided for providing a tutorial message to a driver in a vehicle using an in-vehicle tutorial system including a multimodal user interface. The method includes assigning an area of improvement, including at least one tutorial task, to the tutorial system, identifying a driving context for which a tutorial task is to be performed, based on the tutorial task and the driving context, selecting a tutorial message encouraging a desired driving behavior, and providing the selected tutorial message to the driver using the multimodal user interface.

A fitness apparatus which allows for simultaneous strength training and core muscle building/strengthening, the apparatus incorporating an inflatable exercise ball. The apparatus includes a base which allows for selectively controlling the lateral movement and compressibility of the ball while performing various exercise routines. Also incorporated in the base are adjustable resistance training devices which allow for a variety of strength training, the resistance training devices allowing routines ranging from very light to sufficiently strenuous to challenge advanced users. The apparatus is collapsible and stowable, and may be used without the exercise ball to facilitate both strength/flexibility training, as well as aerobics.

The invention belongs to the technical field of image segmentation, in particular to an image segmentation method, a system and an electronic device based on depth learning. The image segmentation method based on depth learning comprises the following steps: step A: normalizing the original image; B, inputting the normalized image into a ResUNet network model, extracting a feature map containing global semantic information from the input image by the ResUNet network model, upsampling the feature map and stacking the feature map to obtain a final feature map; step c: classifying pixel by pixelthe feature map after upsampling and stacking processing, and outputting the image segmentation result. The present application solves the common problem of gradient disappearance in a convolution neural network, and also enables the network to be more easily trained and converged to a better segmentation result.

The invention relates to the technical field of biology; in order to solve the deficiency of wild phellinus igniarius resources at present, and the problems of insufficient medium nutrition supply, small sporophore growth, and easy undesired microbe infection of bag cultivation in artificial cultivation, the invention provides a phellinus igniarius bag cultivation medium and a method for cultivating phellinus igniarius sporophores by using the medium; the phellinus igniarius bag cultivation medium is prepared by mixing nutrients with water, packaging the mixture in bags, sterilizing the bags at a temperature of 110-135 DEG C, at a pressure of 0.05-0.25 MPa, for 30-180 min, wherein the mass ratio of the nutrients to water is 1:1.8-2.5. With the phellinus igniarius bag cultivation medium formula of the invention, nutrients absorbed during phellinus igniarius sporophore growth are provided, and the cultivated phellinus igniarius sporophores are similar to wild phellinus igniarius with stable quality; and the cultivation mode of the invention is simple, and can be used for large scale cultivation of phellinus igniarius sporophores with medicinal value.

The invention discloses a method for selecting a preset value of an image-guided ultrasonic diagnostic apparatus. Information about general information of the preset value, a source preset value and modification parameters is automatically and associatively stored by fully utilizing important information about use habits of doctors, historical check images of patients and the like when the doctors store an image; and the information is visually displayed to the doctors through various user interfaces so as to rapidly acquire the preset value expected by the doctors under the guidance of the image information. The method greatly reduces the difficulty of adjusting the parameters of the preset value and the doorsill for the use of the ultrasonic diagnostic apparatus, reduces workloads of the doctors for manually adjusting the parameters, and enables less experienced doctors to rapidly adjust the optimized image quality suitable for the patients and visually learn the parameter adjusting experience from images adjusted by the experienced doctors so as to shorten own learning process.

The invention discloses an image white balance method based on a convolutional neural network. The method is applied to a computing device. The method comprises the steps of collecting multiple real scene original images with reference colors in different scenes, and extracting two-dimensional chromatic value histograms of the real scene original images and to-be-processed images, wherein the chromatic values comprise two of R gain, B gain and G gain; obtaining ideal correction parameters set when a user corrects the real scene original images; designing the convolutional neural network for carrying out image white balance correction; training the convolutional neural network according to the two-dimensional chromatic value histograms of the real scene original images and the ideal correction parameters; and inputting the two-dimensional chromatic value histograms of the to-be-processed images into the trained convolutional neural network, thereby obtaining the ideal correction parameters of the images, and correcting the images according to the ideal correction parameters. The invention also discloses an image white balance device based on the convolutional neural network and the computing device.

The invention discloses a spatial-domain image steganography method and system based on generative adversarial network. The carrier image is converted into a probability graph through the generation network of a U-shaped structure, and then the probability graph is coded by utilizing a hyperbolic tangent coding module, a tampering point graph is generated, and the carrier image and the tampering point graph are added to generate a secret-carrying image; and then a steganography analysis network is used for distinguishing the carrier image and the secret-carrying image, and the classification result is fed back to the generation network in an error mode; and finally, the trained generation network and the coding module are combined together, as a final spatial-domain image steganography model, the carrier image is input into the whole model, and the secret-carrying image is output. The invention further discloses a space-domain image steganography system based on the generative adversarial network, and the system includes a generation network module, an encoding module and an image steganography module. According to the spatial-domain image steganography method based on the generative adversarial network, the security is obviously improved, and the design is simple.

The invention discloses a face recognition method based on distinguishability feature fusion, comprising the following steps: A, cutting a global image and at least two local images in each training sample image; B, carrying out model training on each intercepted image by adopting a multi-loss function to obtain a corresponding model, wherein the multi-loss function is obtained by combining the a-softmax loss function with the center loss function. C, carrying out fusion and dimension reduction of each model obtain by training by using a ternary loss triplet loss function, and obtaining the final depth feature of the training sample image. The technical proposal disclosed in the application can solve the problem of data fusion, face posture and model fusion in the process of face recognition by using CNN, and achieve better face recognition effect.

This invention relates to systems to regulate pace. More particularly, this invention relates to providing a user with a regulated pace through the use of an electronic pace regulating, timing, and coaching device. In accordance with the instant invention, a portable programmable electronic device is configured to provide an audible signal at a repeated frequency that is selectable by the user. The pace timer and coaching device provides the user with a set audible cue at a set frequency to be followed during an exercise session. The audible cues provide a single regulated pace to be followed by the user while performing repetitive actions, motions, or strokes such that a desired training or racing pace is adhered to. The device is configured to detachably couple to swim goggles, to a user, to sunglasses, to an arm band, to exercise equipment, or to other articles of clothing utilized while exercising.

The invention discloses an antibiotic Bacillus amyloliquefaciens with powerful bacteriostasis and wide bactericidal spectrum. The strain has been preserved in the China General Microbiological Culture Collection Center and the taxonomy name of the strain is: Bacillus amyloliquefaciens with preservation number CGMCCNo.4777. The antibiotic Bacillus amyloliquefaciens has powerful bacteriostasis, wide bactericidal spectrum, high inhibition rate up to above 93.5% on the rice blast fungus, the Botrytis cinerea and the Phytophthora parasitica; the Bacillus amyloliquefaciens also has very strong stress resistance, and very strong resistance to the acid and alkali, and is suitable to be used as microbial bactericide.

A method and apparatus for automatic detection and labeling of 3D spinal geometry is disclosed. Cervical, thoracic, and lumbar spine regions are detected in a 3D image. Intervertebral disk candidates are detected in each of the spine regions using iterative marginal space learning (MSL). Using a global probabilistic spine model, a separate one of the intervertebral disk candidates is selected for each of a plurality of labeled intervertebral disk locations.

The invention discloses a method for producing recombinant human-like collagen by expression by Pichia pastoris, which adopts Pichia pastoris C13 with a collection number CGMCC No.4437 as a strain and comprises: culturing primary seeds, culturing secondary seeds, culturing in a fermentation tank and inducing expression; and centrifuging fermentation liquor obtained after induced expression, collecting supernate after centrifugation, and performing concentration, gel column chromatography, precipitate decolorization, ion-exchange column chromatography and ultrafiltration for desalination in turn to obtain the recombinant human-like collagen. Compared with other purification methods, the method is simpler in process and higher in target protein yield, realizes a recovery rate of over 60 percent and a purity of over 95 percent, can realize complete decolorization and is suitable for industrial large-scale production of the recombinant human-like collagen.

The invention provides an SVM (support vector machine) classifier training sample acquiring method, a training method and a training system. The SVM classifier training sample acquiring method includes calculating to acquire distance of each sample of an SVM classifier; according to the distance of each sample, clustering the samples for the first time to acquire at least one first category and the samples contained by each first category; clustering the samples for the second time to acquire at least one second category and the samples contained by each second category, wherein a second distance threshold value is larger than a first distance threshold value; dividing the samples in the second categories containing one sample only into isolated samples; selecting one sample in each first category as a representative sample, and setting the training samples of the SVM classifier according to the representative samples and the isolated samples. By the SVM classifier training sample acquiring method, the training method and the training system, number of the samples can be reduced effectively, complexity of sample space of the classifier is lowered, and classifier training is enabled to be simpler and more effectively.

The invention discloses a method for producing astaxanthin by haematococcus pluvialis. The method comprises the steps of: firstly, naturally culturing the haematococcus pluvialis in water of a culture pool for 6-8 days so that the haematococcus pluvialis proliferates, then adding potassium dihydrogen phosphate and sodium nitrate to the pool water, and covering the pool with a red thin film, so that algae cells of the haematococcus pluvialis rapidly grow to reach conversion state under the irradiation of 630nm-760nm light waves; transferring the algae cells at the conversion state to pool water with potassium dihydrogen phosphate, the final concentration of which is 0.02-0.05g/L, covering the pool with a blue thin film, so that the algae cells of haematococcus pluvialis are cultivated to mature under the irradiation of 430nm-490nm light waves, and thus obtaining haematococcus pluvialis, the algae cells of which are rich in astaxanthin, wherein potassium dihydrogen phosphate, sodium nitrate and red lights can enable the algae cells to rapidly grow and reach conversion state in shorter time, and the potassium dihydrogen phosphate and blue lights can enable the algae cells to rapidly grow, mature and accumulate astaxanthin. The method has the advantages of little investment, cultivation simpleness and high economic benefits.

The invention discloses a novel application of malt oligosaccharide based mycose synthetase and malt oligosaccharide based mycose hydrolase in mycose production. The malt oligosaccharide based mycose synthetase has an amino acid sequence as shown in SEQ ID NO:1; and the malt oligosaccharide based mycose hydrolase has the amino acid sequence as shown in SEQ ID NO:3. The malt oligosaccharide based mycose synthetase and malt oligosaccharide based mycose hydrolase disclosed in the invention also have higher most-suitable reaction temperature and thermal stabilities, and lower most-suitable pH values, so that contamination risks are lowered, production stability is improved, the mycose production efficiency by acting reducing glucidtemns through the malt oligosaccharide based mycose synthetase and malt oligosaccharide based mycose hydrolase and pullulanase in a combined manner is improved and the mycose cost is lowered, therefore, a solid foundation is laid up for the extensive application of the mycose in medicines, foods and cosmetics.

The invention relates to an integrated sewage treatment device with high efficiency and low consumption and a treatment method, and belongs to the field of sewage treatment. The integrated sewage treatment device comprises a pretreatment unit, a multi-oxidation-deoxidation environment coupling biological reaction pond unit and a deep treatment unit sequentially connected. The multi-oxidation-deoxidation environment coupling biological reaction pond unit adopts a three-stage anoxia-aerobiotic alternating reaction pond filled with a circular composite glass ball porous microbial carrier. The deep treatment unit comprises a filter cloth dynamic film biological reaction pond and first and second dynamic films. Sewage is treated through the devices sequentially. The device provided by the invention cannot only remove carbon containing organic matters, but also has good nitrogen and phosphorus removal capacity.

The invention discloses an INS (inertial navigation system)-assisted wireless indoor mobile robot positioning method, belonging to the technical field of wireless positioning of a robot. The positioning method comprises a training stage and a pre-estimating stage. The training stage comprises the following steps of: integrating an INS and a WSN (web shell navigator) in a local relative coordinate system; and carrying out data fusion on obtained synchronized navigation data by expanding a Kalman filtering wave to obtain continuous and stable navigation information. The pre-estimating stage comprises the following steps of: inputting the position and speed information measured by the INS into the training stage, and carrying out error compensation by a neural-network-trained INS error model to obtain optimal navigation information. According to the method provided by the invention, the INS positioning precision can be improved, and the positioning range of the indoor robot can be expanded on the basis that the WSN network scale is reduced.

The invention provides a culture method of salt-tolerant anaerobic ammonium oxidation bacteria, which comprises the following steps of: A, enrichment culture: performing enrichment culture on the strain to obtain anaerobic ammonium oxidation bacteria; and B, bacteria acclimation culture: gradually increasing the concentration of salt in the culture solution to perform salt tolerance acclimation culture of the anaerobic ammonium oxidation bacteria subjected to the enrichment culture in the step A, wherein the acclimation period of the bacterial at each salt concentration is 15-30 days; and the step A and step B both adopt sequencing batch culture, and the dissolved oxygen value in the culture solution is 0-0.5mg/L. The bacteria cultured by the method provided by the invention are applied to high-salt ammonia nitrogen wastewater, and the total nitrogen removal rate is high. By adopting the method provided by the invention to culture the salt-tolerant anaerobic ammonium oxidation bacteria, the range of available substrates is wide, and the culture is easy to implement; and additional organic carbon source is not required in the bacteria culture and application process, investment is greatly saved, and the wastewater treatment cost is reduced. The bacteria culture method provided by the invention can be applied to large-scale production, and has good economic and environmental benefits.

A horizontal and super combined beehive comprises a wide-comb box, a frame support, a powder-releasing cover, a pollen box, a connector, a degummed drape, an inserted frame, an integrated frame and a wax comb foundation with suspension loops. The wide-comb box is characterized in that the both ends of the box are used as the breeding areas for two swarms of bees, a super is supported above the middle production area, frame queen excluders are used to form a working bee centralized production area, the areas are covered; the powder-releasing cover is bottomless, pollens and honey on a powder-leaking railing pedal with anti-theft gantry boxes on both sides directly leak in the pollen box at the bottom; the connected box is coarsely adjusted by a multi-teeth hook and finely adjusted and connected firmly by a screw crow bar through screws; the laps and edgings of the inserted frame are integrated, an upper beam and a lower beam are inserted in the upper and lower grooves of two edgings to integrate, the upper beam is thin, the lower beam and the edgings are provided with grooves to be inserted in the comb foundation; a wide hook with the same width and an adjuster are hung on the wax comb foundation for tightening; the laps are integrated with a frame-distance clip and connected with the upper beam through a downward slope, the tops of comb-forming frame supports are butted to each other and are tightly pressed by the degummed drape and the inner frame of the box cover, the distance of the frames is standard so that people can not stung by bees; and the top of each lap is provided with a handle, thus the laps can be fast gripped outside the box. The wide-comb box can be used to multiply the efficiency, thus being suitable to be used for beekeeping around the world.

The invention relates to the technical field of image processing, in particular to a fetus head full-automatic segmentation method based on three-dimensional ultrasound, which comprises the followingsteps of firstly, carrying out data enhancement on a three-dimensional ultrasound volume data set of a fetus head to obtain an enhanced data set; inputting the enhanced data set into a full convolutional neural network, and training a model in an end-to-end volume mapping manner to realize pre-segmentation of the data set; and finally, carrying out iterative optimization processing on a pre-segmentation result by adopting a cascade full convolutional neural network based on an automatic context to obtain a final segmentation result. The invention aims to overcome a plurality of defects in fetal head measurement by the existing two-dimensional ultrasound, so that the subsequent diagnosis efficiency of doctors is improved, and more other prenatal studies are promoted.

The invention relates to saccharification processing technology of jerusalem artichoke by utilizing high-quality fructose biomass as a raw material and strain selection and technology optimization for producing mannitol by taking jerusalem artichoke as carbon source through fermentation. The method comprises the following steps: 1) crushing jerusalem artichoke tuber into coarse particles, filtering after water leaching and enzymolysis for 6 hours, supernating at 42 DEG C, rotating, evaporating and concentrating to obtain saccharification jerusalem artichoke juice with high concentration of fructose; 2) establishing high performance liquid chromatography analysis and detection conditions which can synchronously analyze the content of fermentation liquor substrate (glucose and fructose) and products (mannitol); and 3) inspecting the capacity for producing lactic acid and mannitol through fermentation by seven lactic acid bacteria by utilizing saccharification jerusalem artichoke juice with different concentration of total sugar, thus determining lactic acid bacteria with high transformation rate and production intensity of fructose, and optimizing production fermentation conditions and the highest concentration of tolerant substrate. Through feed-batch fermentation, production efficiency can be improved and mannitol can be continuously produced in large scale. The method not only generates no byproduct of sorbitol, but also has low production cost, wide raw material sources, simple technology, and mature technical route and can be implemented in industrialization.