12results about How to "Highlight substantive" patented technology

The present invention relates to a capsule locking mechanism of a capsule filling machine, wherein a load-bearing flap is connected to a frame; a tray is mounted on the load-bearing flap through a connecting component, the tray is located on the front of the load-bearing flap, and the tray has a support portion for placing a capsule tray; a locking disk component is arranged on the front of the tray. When the capsule tray is placed on the load-bearing flap, the capsule tray is locked by the tray and the locking disk component, which is convenient for locking the capsule and driving the capsule tray to rotate and unload the capsule, and provides convenience for the subsequent arrangement of a needle tray and a capsule locking baffle. The needle tray only needs to be arranged between the tray and the load-bearing flap and move axially relative to each other. When locking the capsule, the needle tray moves axially toward the front direction, and when locking the capsule, the load-bearing flap is in a horizontal state, and the capsule locking baffle rotates to the outside of the capsule tray to block the capsule tray. The structural layout is reasonable to prevent the loss of medicine powder; after locking the capsule, the capsule locking baffle rotates to give way, and the load-bearing flap is flipped, so that the needle tray moves toward the front to push out the locked capsule in the capsule tray, thereby unloading the capsule.

The invention belongs to the technical field of microorganisms, and particularly relates to a high-yield high-temperature-resistant pectinase strain and application thereof. The invention provides an aspergillus niger mutant strain AR-11 for producing high-yield and high-temperature-resistant alkaline pectinase, the enzyme activity level of pectinase produced by utilizing the strain for industrial fermentation can reach 8000-8200U / mL or more, the production cost is greatly reduced, and large-scale application, energy conservation and emission reduction of the alkaline pectinase are facilitated. The produced pectinase has high enzyme activity under a high temperature condition, keeps the stability of the enzyme activity in a wide pH value range, can meet the requirements of a pectinase preparation in the aspects of ramie degumming and textile refining, and can be widely applied to the ramie degumming industry.

The invention relates to a line patrol unmanned aerial vehicle power line patrol data fusion method based on multi-aerial vehicle cooperative control, provides deep learning, and belongs to the field of image or video recognition or understanding. The method comprises the following steps: a ground cooperative control station synchronously issues an N-view-angle line patrol task of a target line segment of a set power line transmission line to N line patrol unmanned aerial vehicles; and adopting an AI inspection data fusion model of a set power line transmission line to intelligently judge various fault types existing in a target line segment at a preset acquisition moment and positioning information of occurrence positions of the fault types on the basis of multiple analysis data corresponding to the N-view-angle line inspection task. In order to solve the technical problem that the precision, efficiency and comprehensiveness of a power inspection result cannot be improved at the same time in the prior art, a hardware architecture for power line inspection data fusion of the unmanned aerial vehicle for multi-vehicle cooperative control is established, and synchronous intelligent analysis of multiple fault types and occurrence positions thereof is completed by adopting an intelligent mode. Therefore, the technical problem is solved.

The present invention discloses a power transmission structure of a household platform elevator, which is connected and linked with a driving motor and a driving nut. In particular, the power transmission structure includes a transmission belt, a first pulley and an offset shaft assembly, wherein the first pulley is keyed to the output shaft of the driving motor and rotates coaxially, and the offset shaft assembly is integrally mounted with a bearing seat, an offset shaft, an encoder and a second pulley based on an offset bottom plate, and the offset shaft and the output shaft of the driving motor are distributed on both sides of the driving nut and are axially parallel, and the transmission belt is sleeved on the first pulley and the second pulley for closed-loop rotation and the transmission belt is pressed against the two sides of the axial center symmetry of the driving nut to transmit power. The power transmission structure of the present invention is applied, and the pressing position of the transmission belt facing the driving nut is improved by adding an offset shaft assembly while ensuring power transmission, and the external force that causes the axial skewness of the driving nut is completely eliminated, which is beneficial to prolonging the service life and maintenance period of the driving nut, and improving the noise and physical sensation of elevator operation.

The invention discloses a dynamic water penetration test system for an emergency power anchor. The dynamic water penetration test system comprises a movable bed and dynamic water circulation module, a power anchor transmitting module, a power anchor drawing module, a video acquisition module and an auxiliary lighting module, the moving bed and moving water circulation module simulates breach water flow and riverbed boundaries, the power anchor transmitting module is used for forcefully penetrating a power anchor into the moving bed and moving water circulation module, and the video acquisition module is used for acquiring the penetration motion trail of the power anchor. The power anchor drawing module is used for testing tension data of pulling a power anchor out of the moving bed and the moving water circulation module, and the auxiliary lighting module is used for providing uniform light illumination for the moving bed and the moving water circulation module. According to the invention, the penetration emission module is taken as a core, a large circulating water tank is taken as a basis, technologies such as image recognition and multi-working-condition cold emission are utilized, two functions of dynamic water penetration and anchor body drawing are integrated, and real-time observation and research on an anchor body penetration process, an interface effect, anchor body offset and an anchoring force test are realized in a dynamic water environment.

This invention relates to a machine vision-based system and method for recognizing and assessing gait abnormalities in children, belonging to the field of action recognition, and more specifically to the field of recognizing human actions or behaviors in image or video data. The system includes: a model building device for building an AI gait abnormality recognition and assessment model corresponding to a side-view imaging device; and an intelligent assessment device for intelligently assessing, based on selectively chosen multi-source basic data, whether the child currently being assessed has gait / crawling abnormalities and the number of months of such abnormalities. This invention addresses the technical problem of simultaneously failing to balance the quality and efficiency of gait / crawling abnormality assessment and the inability to simultaneously provide the number of months of abnormalities. It employs a custom-designed AI gait abnormality recognition and assessment model, based on selectively chosen multi-source basic data, to simultaneously complete both the status assessment and the assessment of the number of months of abnormalities, thereby solving the aforementioned technical problems.

The application discloses a chemical precipitation method using magnesium oxide and a metal hydroxide product, and the chemical precipitation method comprises the following steps: S1, obtaining a pre-reaction solution capable of being used for a precipitation reaction, and obtaining a pulp solution containing ions of a target metal in the pre-reaction solution; S2, mixing magnesium oxide powder and the pulp solution, and obtaining a first slurry through a pulp treatment; and S3, putting the first slurry into a reactor to perform a precipitation reaction, and generating a precipitated slurry containing a metal hydroxide of the target metal. In the magnesium oxide powder, a first part of the magnesium oxide powder has a higher average activity than that of the whole magnesium oxide powder. The method is used to solve the problem of serious activity attenuation of the magnesium oxide slurry, improve the precipitation efficiency of the target metal in the precipitation reaction, increase the content of the target metal in the metal hydroxide product, and reduce the single consumption of the magnesium oxide and the content of magnesium impurities in the product.

The present invention provides a training sample image enhancement method and system based on a neural network, the method comprising the following steps: manually labeling sample images in an initial sample set to obtain semantic segmentation labels; using the semantic segmentation labels to segment the sample images, and generating multiple combination information according to the similarity of the segmented images; calculating the preference of each combination information, and selecting a pre-expanded number of combination information as an augmented image in descending order of preference; using a training discriminator to judge the authenticity of the augmented image, and optimizing the image fusion parameters according to the authenticity of the augmented image to obtain an augmented sample set. The present invention amplifies more similar images that meet the requirements by splicing and fusion methods such as the interchange of foreground objects and backgrounds in different sample images, which not only increases the quantity and richness of sample data, but also reduces the corresponding manual labeling workload.

The application discloses a PCB defect image detection method based on an improved deep learning algorithm. In view of the deficiencies of the YOLOv3 deep convolutional neural network model, such as imbalance of positive and negative samples, feature expression of small defects of the PCB, and extraction of important feature information of the image, the application proposes an improvement method of replacing the MSELoss function in the original YOLOv3 model with the FocalLoss function, so as to dynamically control the weights of positive and negative samples and difficult and easy classification samples. Meanwhile, the application embeds the SPP module in the branch small target detection head and parallelly integrates the CBAM module into the YOLOv3 model, so that the model can pay more attention to the defect information in the image, thereby improving the overall feature extraction capability of the model. The method can detect different types of PCB defect images, and compared with the original YOLOv3 model, the sensitivity and other indicators are improved, has good performance in the PCB defect detection task, can reduce the missed detection rate and the false detection rate, and better realizes the automatic and intelligent detection of the PCB defects.

The invention provides a low-temperature vacuum extraction system for soil or plant moisture, which is characterized in that a pre-cooling / heating table, a refrigeration table and a plurality of rod core type pipelines are arranged in automatic extraction equipment, placement positions corresponding to sample tubes are arranged in the rod core type pipelines, and collection test tubes are detachably arranged at the lower parts of the rod core type pipelines; a placement through hole is formed in the pre-cooling / heating table, a pre-cooling pipeline and a heating pipe are wound in the pre-cooling / heating table and correspond to the middle of the rod core type pipeline, a groove is formed in the refrigerating table, and a refrigerating pipeline is wound in the refrigerating table and corresponds to the groove; the refrigerant source is communicated with the pre-cooling pipeline and the refrigerating pipeline; a lifting platform is arranged at the upper part of the automatic extraction equipment, a vacuumizing system is arranged in the lifting platform, a floating end cover is arranged at the bottom of the lifting platform corresponding to a top port of the rod core type pipeline, and the floating end cover is communicated with the vacuumizing system; a control unit is arranged on the automatic extraction equipment and is in control connection with the vertical lifting mechanism and the heating pipe. The method has the advantages of high efficiency, low leakage rate, safety and convenience.

This invention provides a controllable image description method, system, device, and medium based on multimodal fusion. The method includes the following steps: based on the original image features X... ν * and object regression box features X ν b The images are fused to generate sample image features X. ν Extract each word from the labeled language text, and label and positionally encode the extracted words to generate sample text description features X. w Generate a trajectory regression box based on the mouse trajectory segment, and generate sample trajectory features X based on the trajectory regression box. r Using a pre-built TO-Transformer feature fusion model, the sample image features X are processed. ν The sample text description feature X w and the sample trajectory feature X r Feature fusion is performed to generate sample fusion features; based on the sample fusion features, a controllable image description model is generated using a pre-built output model, thereby effectively improving the controllability and efficiency of image description.

A high-yield and high-efficiency drip irrigation method for cotton cultivation based on root system configuration time-series optimization is proposed. Target root system configurations are set for three stages: seedling stage, budding to early flowering stage, and boll-forming to boll-opening stage. A coupled model of "drip flow rate - water spatial distribution - root morphology response" is used to inversely design the root zone water field at each stage, thereby dynamically configuring dripper flow rate and irrigation parameters. A dripper flow rate of 0.8–1.1 L / h is used during the seedling stage to induce the formation of slender, spreading roots; a dripper flow rate of 1.5–1.8 L / h is used during the budding to early flowering stage to promote root thickening; and a dripper flow rate of 2.2–2.4 L / h is used during the boll-forming to boll-opening stage to ensure high yield and quality. Field trials show that compared with a fixed low flow rate, yield is increased by more than 120%, and irrigation water productivity is increased by more than 20% compared with a fixed high flow rate, achieving a synergistic effect of high yield, high efficiency, and high quality in cotton cultivation.