7results about How to "Avoid neglect" patented technology

The present application relates to the technical field of auxiliary identification positioning, in particular to an AI-assisted lesion positioning method and system based on a gastrointestinal endoscope. Based on segmentation boxes of different sizes, a gastric ulcer lesion area and a corresponding confidence level in a gastrointestinal endoscope image are identified. The completeness value of the gastric ulcer lesion area is determined. According to the number of times that a pixel point at each position is divided into the gastric ulcer lesion area under segmentation boxes of different sizes, in combination with the confidence level and the completeness value of the gastric ulcer lesion area to which the pixel point is divided, the region belonging value of each pixel point is obtained. Based on the region belonging value, the pixel points are merged to obtain a preliminary screening lesion area. The region belonging value of the pixel points and the distribution of the region in the preliminary screening lesion area are analyzed to determine the performance bias degree of each preliminary screening lesion area. Based on the performance bias degree, a target gastric ulcer area is determined from the preliminary screening lesion area. The present application improves the identification accuracy of the target gastric ulcer area.

This utility model discloses a construction equipment for building composite insulation boards, belonging to the technical field of construction equipment. The equipment includes a trolley frame with a screw rotatably connected to it. A slide rod is fixedly connected to the trolley frame at the side of the screw. A first positioning frame is movably mounted on the screw and slide rod, and a second positioning frame is rotatably connected to the side of the first positioning frame. The equipment uses a stepper motor to drive the screw to rotate, making it easy to change the height of the first and second positioning frames and adjust the included angle between them. This facilitates the clamping and fixing of the insulation board at the corner of the exterior wall, improving practicality. Furthermore, the suction cups on both sides of the clamping board adhere to and fix the insulation board, resulting in better fixation. When the insulation board slides downwards, it separates from the suction cups, making it easy to detect whether the insulation board has moved, preventing workers from overlooking it and allowing for timely adjustment of the insulation board's position.

This application provides an adaptive integrated land surface classification method and system based on multi-scale spatiotemporal partitioning, belonging to the field of remote sensing image processing technology. The method includes: acquiring multi-source remote sensing image data, digital elevation model (DEM) data, and historical land use data; partitioning the target area into spatial units based on the DEM data; identifying the phenological change process of each spatial unit using time-series data; determining spatiotemporal partitions based on the change process, historical land use data, and each spatial unit; constructing a feature set based on the spectral, textural, and index features of the topography and phenological characteristics of each spatiotemporal partition; selecting and applying a corresponding pre-trained land surface classification model based on the feature set and spatial units to obtain the classification result; and establishing a feedback update mechanism based on the classification result, multi-source remote sensing image data, and DEM data. This application improves classification efficiency and accuracy.

The invention discloses an equipment control method and system based on deep learning, and relates to the technical field of compressor control, and the method comprises the steps: obtaining a historical normal sequence and a current operation sequence, and carrying out the preprocessing to obtain a historical feature data set and a current feature data set; calculating the attention weight of each feature under each time scale based on the historical feature data set; dividing the historical feature data set and the current feature data set to obtain a historical fusion window and a current fusion window under each time scale; combining the attention weight to calculate a fusion value and construct a fusion sample, and obtaining a historical fusion sample set and a current fusion sample under each time scale; constructing and training a prediction model to obtain an effective prediction model under each time scale; and calculating a comprehensive valve opening value at the next sampling moment based on the current fusion sample and the effective prediction model. According to the method, the valve opening of the compressor can be accurately predicted, and the high-precision control requirement of the compressor under the complex working condition is met.

This utility model provides a roller surface cleaning mechanism for an electronic belt scale. It relates to the field of electronic belt scale cleaning technology. The roller surface cleaning mechanism includes a base, on which four mounting brackets are fixedly installed. Each of the four mounting brackets is equipped with a connecting component. Each connecting component includes: two upright blocks, a connecting rod, a circular roller, two upright blocks, two rotating blocks, two connecting rods, two rotating blocks, two circular rollers, and two upright blocks. Two upright blocks are fixedly installed on the top of their respective mounting brackets. The connecting rod is fixedly installed between the two upright blocks. The circular roller is rotatably mounted on the connecting rod. The two upright blocks are fixedly installed on the top of their respective mounting brackets. This utility model has the advantages of accelerating cleaning efficiency, reducing time and effort spent, and relatively improving cleaning efficiency.

The invention discloses a rocket artificial rainfall enhancement influence area estimation method and system. The method comprises the steps of A1, collecting rocket rainfall enhancement operation parameter data, regional wind field data and atmospheric water vapor distribution data, and performing preprocessing respectively; a2, respectively extracting structured diffusion characteristics, regional wind field characteristics and water vapor distribution characteristics; a3, carrying out cross-modal fusion to obtain a multi-modal fusion feature and a catalyst diffusion feature; a4, extracting catalyst-water vapor interaction characteristics; a5, calculating a basic influence area and a basic grid range image of the rocket artificial precipitation enhancement operation; and A6, performing grid statistics and proportional scale conversion on the accurate influence area of the artificial precipitation enhancement operation of the rocket to obtain the accurate influence area of the artificial precipitation enhancement operation of the rocket. According to the method, the problem that the estimation accuracy of the rocket artificial precipitation influence area is insufficient due to the fact that linear source diffusion is simplified into a trapezoidal model and multiple factors are not fully considered in a traditional method can be solved.

The application discloses a human-computer collaborative intelligent interaction state tracking method and system, relates to the technical field of human-computer collaborative interaction, and comprises the following steps: acquiring first feature data and second feature data; extracting amplitude information of the first feature data, and generating a first identifier and a first parameter according to whether the amplitude information is lower than a first threshold value; in the case that the first identifier is captured, third feature data of the second feature data is extracted; based on a second identifier of a terminal, the first parameter and the third feature data, target score information is determined; in the case that the target score information satisfies a second threshold value, a target signal is issued to an external collaborative device, and an asynchronous state backtracking check is triggered; the application avoids the emergency interaction instruction of the cloud traditional Chinese medicine platform being ignored in an extreme care scene.