41results about How to "Realize automatic grading" patented technology

The invention belongs to the technical field of medical image processing and computer vision, and particularly relates to an MRI image liver fibrosis automatic grading method based on imaging omics analysis. The invention particularly relates to an imaging omics analysis method for grading magnetic resonance dynamic enhanced imaging liver fibrosis of a Gd-EOB-DTPA contrast agent. The method comprises the following steps: establishing an initial data set by taking a Gd-EOB-DTPA dynamic enhanced image of a hepatitis B patient as source data; registering the initial data; carrying out automatic liver segmentation by utilizing transfer learning to obtain an ROI; performing image omics analysis: for the ROI, carrying out image omics feature extraction and feature screening on a plurality of DCEsequences to obtain a feature subset of important features, and carrying out training of a plurality of classifiers to select an optimal classifier; and finally, predicting and evaluating the classification performance in an external test set by using the classifier. According to the method, the accuracy and reliability of automatic liver fibrosis grading based on DCE images can be effectively improved, and meanwhile, the time and energy of clinicians are saved to a great extent.

The invention discloses a hepatocellular carcinoma automatic grading method based on an SE-DenseNet deep learning framework and an enhanced MR image. The method comprises the following steps of 1) collecting data; 2) preprocessing all hepatocellular carcinoma three-dimensional images with enhanced MR; 3) enhancing the training data; 4) based on the enhanced training data, training a hepatocellularcarcinoma grading prediction model, namely an SE-DenseNet network; and 5) carrying out hierarchical prediction on the test data by adopting the trained model, and evaluating the classification performance of the hepatocellular carcinoma hierarchical prediction model. According to the automatic pathological grading method for the hepatocellular carcinoma multi-modal enhanced MR image which is composed of the steps of image preprocessing, the image enhancement, the hepatocellular carcinoma multi-modal enhanced MR image classification, SE-DenseNet network training and SE-DenseNet network testing, the hepatocellular carcinoma automatic grading can be realized, and the problems of manpower consumption, time consumption and subjective difference existing in manual hepatocellular carcinoma grading can be solved.

The invention discloses a method of granulation by spraying melted nicotinamide. The method comprises the following steps: injecting the melted nicotinamide into the spray-granulation tower of a fluidized bed for granulation by a pressure-type spraying nozzle; automatically grading the nicotinamide granules with the granularity being 0.4 to 1.0 mm in the granulation tower by controlling the gas velocity of fluidizing air in the inlet pipe at the lower part of the granulation tower; entering silos through the inlet pipe of the granulation tower of the fluidized bed, and then screening the products within the range of the granularity by a screener; and finally obtaining the nicotinamide granules with the granule size being 0.4 to 1.0 mm. On the one hand, the granular nicotinamide produced by the method of the invention can greatly reduce the dust pollution, moisture absorption and agglomeration occurring to places for packaging, storing, transporting and applying, and on the other hand, the granule sizes of the products are properly distributed so as to ensure the good dispersion of nicotinamide in the following downstream products, such as feed and the like, upon the premise that the following processing technique, such as feed and the like, remains unchanged.

A fruit classification system with mechanical vision controlled by shift register is composed of an external power unit, a delivery unit and a classifying unit which are composed of two chain wheels and chain, an encoder installed on the delivery unit, an image acquisition card in computer, an illuminating box, a camera in the illuminating box, a computer, a shift register set connected to the computer and the classifying unit, a discharge port, and a software. Its advantages are no destruction to fruit.

The invention discloses a part three-dimensional dimension detection machine, which comprises a feeding rack and a main machine body. The feeding rack is provided with a vibration feeding mechanism. The main machine body is internally provided with a main base. The main base is provided with a feeding pusher mechanism, a dividing plate, a discharging pusher mechanism, a detection mechanism and a sorting mechanism. The feeding pusher mechanism can push parts fed by the vibration feeding mechanism to the dividing plate. The discharging pusher mechanism comprises a discharging pusher block and a discharging pusher driving mechanism. The detection mechanism comprises a detection platform, an upper dimension detection sensor and a lower dimension detection sensor. The discharging pusher block can push the parts on the dividing plate to the detection platform. The upper dimension detection sensor and the lower dimension detection sensor are used for detecting the three-dimensional dimensions of the parts. The sorting mechanism comprises a conveyor belt, a conveyor belt drive motor, a side sorting hopper and a sorting blowing device. The detection machine can realize continuous material-pushing detection, is high in work efficiency and high in detection precision of the three-dimensional dimensions of the parts, and meanwhile, can realize automatic sorting of the parts.

The invention discloses a geological detection method and system based on a blower gun seismic source. The method comprises the steps of: building a measurement network at a position which is a certain distance away from a tunnel face, and setting a measurement point for drilling on the measurement network; acquiring a seismic wave signal generated by a blower gun as an excitation source at the measurement point, and performing advanced geological forecast by adopting an inversion imaging model based on the seismic wave signal to obtain an unfavorable geologic body positioning result; acquiring working parameters of the blower gun in the drilling process, and obtaining a surrounding rock grading result based on the working parameters through a classification model; and controlling the tunnel construction process based on the unfavorable geologic body positioning result and the surrounding rock grading result. A blower gun is used as an excitation source for advanced geological forecast, and an inversion imaging model is repeatedly corrected through seismic wave signals, so that fine detection of unfavorable geology of the tunnel is realized; and meanwhile, the working parameters of the blower gun in the drilling process are collected, automatic grading of the surrounding rock is achieved based on a support vector machine algorithm, and finally tunnel construction is effectively guided according to geological detection results in the two aspects.

A Gordon Euryale kernel classifier comprises a feeding hopper, screening machines and a grit screening machine. A discharging barrel is arranged at the lower end of the feeding hopper. The grit screening machine is arranged below the discharging barrel and comprises a base IV, supporting plates, a collecting plate and a conveying belt. The supporting plates are arranged at four corners of the baseIV respectively, a roll is arranged between every two supporting plates at the same end, and the conveying belt is arranged on the two rolls. The conveying belt inclines towards one side, the supporting plates at the lower side of the conveying belt are provided with the collecting plate inclining downwards, and inclined baffles are arranged on two sides of the collecting plate. The screening machines are arranged below a free end of the collecting plate and each comprise a roller screen mechanism, a motor and a base. Each roller screen mechanism comprises supporting pillars, an inner-layer belt, an outer-layer belt and drums, wherein the outer-layer belt is provided with a plurality of strip-shaped holes. The Gordon Euryale kernel classifier can realize automatic grading of Gordon Euryale kernels, broken Gordon Euryale kernels can be automatically separated out, time and labor are saved, and the work efficiency is improved.

The invention discloses a grassland degeneration grading method based on a bp neural network, and relates to a grassland degeneration degree evaluation method, which comprises the following steps: S1, determining influence factors of grassland degeneration, and obtaining original data related to the influence factors; s2, processing the original data related to the influence factors to obtain standardized data; s3, importing the standardized data into a pre-trained bp neural network model, and outputting an evaluation grading result; and S4, translating and transcoding the evaluation grading result, and outputting the result in a friendly manner. The beneficial effects of the invention are as follows: the grassland degeneration is automatically graded based on the bp neural network according to a plurality of collected data by taking five main factors which affect the grassland, namely the coverage degree of convex spots, the proportion of dominant grass seeds, the proportion of degeneration indicating seeds, the proportion of edible pasture and the rat damage condition as entry points, and the grassland degeneration grading method can be widely applied to the aspects of grassland degeneration research and the like.

The invention discloses a fruit grade selecting and packing weighing device and system and a treatment method, and belongs to the technical field of fruit treatment equipment. The device comprises a sorting part, a basket packing part and a controlling part; the sorting part comprises a fruit conveying belt, a high-quality fruit conveying belt and a transferring conveying belt, the fruit conveyingbelt is protruded from one or two sides of the high-quality fruit conveying belt relatively, and a buffer anti-drop structure is arranged on the transferring conveying belt; and the basket packing part comprises a first fruit basket conveying belt and a second first fruit basket conveying belt which are arranged in sequence, the first fruit basket conveying belt sends the fruit baskets below thetail end of the fruit conveying belt for primary basket packing, the second fruit basket conveying belt sends the primary-packed fruit baskets to below the buffer anti-drop structure for secondary basket packing, a first position sensor and a second position sensor are arranged at the basket packing positions of the first fruit basket conveying belt and the second first fruit basket conveying beltrespectively, and a first electronic scale and a second electronic scale which are used for detecting the weight of the fruit baskets are arranged at the basket packing positions of the first fruit basket conveying belt and the second first fruit basket conveying belt respectively.

The invention relates to an electromagnetic damping type automatic tobacco sorting manipulator. The manipulator is arranged above a conveying belt used for conveying tobaccos, and comprises a brush mechanism and two guide rails, wherein the brush mechanism is used for sweeping the tobaccos on the conveying belt into a tobacco collecting box, the two guide rails are arranged in parallel up and down, and form a rail which is across the conveying belt to be used for enabling the brush mechanism to move back and forth on the rail, an electromagnet is arranged on one end of the rail, an elastic damping mechanism is arranged on the other end of the rail, the brush mechanism is connected with the elastic damping mechanism, an armature which can overcome the elastic force of the elastic damping mechanism so as to be attracted to the electromagnet when the electromagnet is electrified is fixed on the brush mechanism, a photoelectric sensor is arranged above the conveying belt, and both the photoelectric sensor and the electromagnet are connected with a control mechanism. The electromagnetic damping type automatic tobacco sorting manipulator has the advantages of simple structure and high degree of automation and is favorable for improving the product quality, releasing the labor force and saving the cost.

The invention relates to the technical field of micro-plastic detection, in particular to a device and method for detecting micro-plastic in air, and the device comprises an environmental data monitoring module, an atmosphere sample collection module, a sample processing module and a controller; the environmental data monitoring module comprises a plurality of environment monitoring sensors used for detecting environment data, and the output ends of the environment monitoring sensors are connected with the input end of the controller. The output end of the controller is connected with the input end of the atmosphere sample collection module and is used for controlling the atmosphere sample collection module to work for sample collection, and the output end of the controller is connected with the input end of the sample processing module and is used for controlling the sample processing module to preprocess a sample. The device for detecting the micro-plastics in the air can realize selective sampling of the atmosphere so as to obtain a required atmosphere sample, and meanwhile, the micro-plastic particles can be classified and screened, so that the detection accuracy is improved.

The invention provides a data security grading and classification method based on deep learning, and relates to the technical field of data security grading and classification, and the method comprises the following steps: 1, collecting business data, obtaining a plurality of groups of security data flow direction sequence combinations, and carrying out grading and marking; 2, extracting flow direction sequence combination features of the security data flow; 3, constructing a data security grading mature model based on deep learning; and 4, performing real-time dynamic monitoring on internal and external data streams which are occurring in the service information system by using the data security grading model, extracting a complete/incomplete dynamic data information stream sequence group for classification, and forming a data security grading result of the A core data, the B important data and the C general data until the group of operation behaviors is finished. Through the method provided by the invention, automatic grading of the service information system data can be realized, real-time grading and scene classification are carried out on mass data with mass data categories in real time, and the working efficiency of a data administrator is improved.

The invention belongs to the technical field of oral cavity image intelligent diagnosis, and particularly relates to a caries automatic grading method and system based on a panoramic film. The method comprises the following steps: S1, segmenting and cutting a tooth instance of a panoramic film to obtain an instance segmentation sample of a single tooth in the panoramic film; s2, using the trained tooth coarse classification model to segment the obtained tooth instance into samples, and performing restoration state and non-restoration state classification; s3, after the tooth samples in the non-repair state pass through the trained tooth tissue semantic segmentation model, a plurality of different regional tissues of the segmented teeth are obtained; and S4, according to the obtained tissue conditions of different areas of each tooth, dividing the tooth into four grades of healthy, shallow caries, medium caries and deep caries through a caries grading model according to caries grade diagnosis standards. The system and the method have the characteristics that classification of tooth health states and automatic classification of caries can be realized, and the workload of clinicians is reduced.