36results about How to "Simplified throughput" patented technology

The invention discloses a tobacco laser code auxiliary identification method, device and tobacco laser code identification equipment. The tobacco laser code assisted identification method includes the following steps: S10, adding a segmentation mark between the custom characters of the tobacco laser code of the cigarette and the number after the digital serial number after the custom characters; S20, obtaining all the The tobacco laser code of the cigarette and the image of the segmentation mark; S30, identify the digital serial number in the image in a preset order, and when the position of the custom character is recognized, use the segmentation mark as The starting position, identifying the number after the user-defined character in the digital serial number. The tobacco laser code assisted recognition method, device and tobacco laser code recognition equipment of the present invention are simple and accurate for tobacco laser code recognition, simplifies the amount of data processing, and can meet the requirements for recognition efficiency and accuracy when a large number of tobacco laser codes need to be recognized .

The present invention provides a kind of Go automatic notation method based on video recognition technology, comprises the following general steps: 1) initialization, earlier obtains a frame to comprise the original image of whole blank chessboard and additional four-corner positioning mark by camera as original frame; 2) Chessboard positioning, using the four-corner positioning method of the back-shaped positioning mark, performing chessboard positioning on the original frame, and obtaining the specific position coordinates of 361 points on the Go chessboard in the image; 3), frame calculation, the video obtained by the camera according to the following The speed of 10 frames per second is taken out, image processing is carried out one by one, and the method of judging the frame difference is adopted to judge the change state of the chessboard points; 4) generate the chess record file. This method can effectively remove the irrelevant image interference around the chessboard, which greatly simplifies the amount of image processing; using the matrix method for judging the frame difference, the image after four-corner positioning and cutting can judge the position of the chessboard by judging the frame difference of the entire picture state.

The invention relates to a method for extracting ursolic acid from privet leaves. The pulverized privet leaves are refluxed with ethanol to obtain a liquid extract, and the ethanol is recovered. The above extracts are fully washed with water and then dried to obtain dry Extract, dry extract is decolorized after dissolving with ethanol, after recovering ethanol, add water to precipitate a large number of precipitates, adjust the pH value to 2-2.5, separate and collect the precipitate, wash with water until neutral, and dry in vacuum to obtain the main components Crude extract of ursolic acid. The present invention adopts the above-mentioned method, and takes the cheap and easy-to-obtain Jurchen leaves as raw materials and extracts them under reflux with ethanol. The amount of "three wastes" is small and easy to handle, ethanol can be recycled and reused, and the residue left after extraction can be used as fertilizer. Therefore, the present invention basically does not pollute the environment, and the cost of the obtained ursolic acid is relatively low. In addition, the method of the present invention is simple and fast.

The present application relates to a data acceleration transmission method, device, computer equipment, storage medium and computer program product. The method includes: based on the routing table configured for the application to be accelerated, routing the data packet sent by the application client corresponding to the application to be accelerated to the virtual network device, reading the network layer data packet from the virtual network device, and analyzing the original destination address and In the data segment, the address of the acceleration server corresponding to the application client is used as the update destination address of the network layer data packet, and the original destination address is added to the head of the data segment to obtain an update data packet, which is sent to the acceleration server so that the acceleration server According to the original destination address corresponding to the head of the data segment, the interaction with the application server corresponding to the original destination address is established, and the response data packet fed back by the application server is returned to the application client, effectively improving the effect of accelerated data transmission. The above method can also be applied to smart transportation to realize accelerated transmission of map data.

A method for predicting the properties of crude oil from near-infrared spectra, including collecting different types of crude oil samples, measuring their near-infrared spectra and physical property data, performing principal component analysis after performing second-order differential processing on near-infrared spectra, and taking the first w scores and After the four physical property data are standardized, the principal component analysis is performed, and the first w+1 scores are taken to establish a crude oil identification database. After processing the near-infrared spectrum of the crude oil sample to be tested, the same identification analysis as the database construction is carried out. The identification variable of the crude oil to be tested is obtained, and the crude oil sample closest to it in the database is determined by the identification variable, and the spectra of these crude oil samples are used for spectral fitting of the crude oil sample to be tested, and then the fitting is judged by comparing the fitting degree with the threshold , for a fully fitted sample to be tested, predict the physical property data of the crude oil sample to be tested from the physical property data corresponding to the library spectrum involved in the fitting.

The invention provides a chessboard positioning method based on video recognition technology for automatic Go notation, which is a chessboard positioning method included in the image recognition-based Go notation processing method, and the chessboard positioning method is based on four corner positioning marks. The glyph icon is a positioning mark, and the specific coordinates of 361 points in the chessboard are calculated according to the positions of the four corners of the chessboard and the horizontal and vertical spacing standards of the chessboard. Arranging back-shaped icons at the four corners of the chessboard as the image recognition control points for chessboard positioning can calculate the specific coordinates of 361 points in the chessboard according to the four corners of the chessboard, and can effectively remove irrelevant image interference around the chessboard, greatly simplifying the image processing.