43results about How to "The solution is not accurate enough" patented technology

The invention discloses a duplication elimination method based on search results of a metasearch engine. The duplication elimination method based on the search results of the metasearch engine is provided to solve the problems that in the prior art, duplication elimination cannot be carried out on the same web pages of URLs with different formats and redirected web pages, and sentences and articles cannot be represented fully according to information around punctuation marks, positions of the punctuation marks and fuzzy matching of word frequencies. The method comprises following steps such as 1, judging whether there are duplicated web pages or not according to URL addresses; 2, computing the title similarity of the two web pages and the abstract similarity of the two web pages; 3, computing the similarity of the web pages according to the title similarity and the abstract similarity; and 4, determining that the two web pages are duplicated if a similarity value Sim (A, B) is more than a threshold. The method can be applied to the field of duplication elimination based on the search results of the metasearch engine.

The invention discloses a battery state of charge SOC correction method and device and an electric automobile, and belongs to the technical field of electronics. The method comprises the steps that the initial state of charge SOC and the first temperature value of a battery pack system at the ith moment are acquired, wherein i>0; the actual SOC of the battery pack system is determined according to the first temperature value; the theoretical SOC of the ith moment is calculated according to the initial SOC and the first temperature value through an ampere-hour integral method; and the theoretical SOC is corrected according to the difference value between the theoretical SOC and the actual SOC so that the corrected SOC corresponding to the ith moment is obtained. The problem of inaccuracy of the calculated SOC caused by the fact that the calculated theoretical SOC is directly used as the real SOC of the battery pack system when the SOC is calculated by a BMS according to the ampere-hour integral method can be solved so that the accuracy of the SOC obtained by the BMS can be enhanced.

The invention discloses a channel quality information computation method, device and system, and belongs to the field of the communication. The method comprises the following steps: a terminal receives a downlink configuration instruction sent by access network equipment; the terminal measures a reference signal source indicated by the downlink configuration instruction to obtain first channel quality information, and sends the first channel quality information to the access network equipment; the terminal sends an SRS to the access network equipment; and the access network equipment computessecond channel quality information according to the first channel quality information and the SRS. Since the first channel quality information can more accurately represent the interference situationof a downlink channel, the problem that the channel quality information of the downlink channel, obtained through the eNB computation is not accurate enough in the prior art is solved; and an effect that the second channel quality information computed by the access network equipment can accurately represent the channel quality of the downlink channel is achieved.

The invention provides a POI coding method, belongs to the technical field of computers, and aims at solving the problem that recommended POIs are not correct enough when POI recommendation is carriedout on the basis of POI coding as the POI coding methods in the prior art do not consider association between the POIs during the coding. The method comprises the following steps of: obtaining a POIsequence of a user operation; training a POI coding model through a deep learning-based neural network architecture according to the POI sequence; and determining a code of each POI in the POI sequence according to a POI coding vector sequence corresponding to the POI coding model. According to the method, POI feature extraction is carried out through the POI sequence generated on the basis of theuser operation, and the POIs are coded on the basis of the extracted features, so that associativity of the POIs is hidden in the POI codes, and then the correctness of carrying out POI recommendation on the basis of the POI codes is further enhanced.

The invention relates to a triaxial planeness detection system and a triaxial planeness detection method. The system comprises four parts including a basic component, a triaxial sliding mechanism, a workpiece locating platform and a laser measurement component. The triaxial sliding mechanism comprises a cross-shaped sliding table and a vertical sliding table, and the cross-shaped sliding table drives a detected workpiece to move in a horizontal range and works in cooperation with a laser measurement instrument. A measurement mode in which the workpiece moves and the laser measurement instrument is stationary is used, thus, accuracy of a measurement result is improved, and a problem that the measurement is not accurate enough when the laser measurement instrument moves and even micro vibration can cause large deviation of the measurement result is solved. The laser measurement component comprises three groups of the laser measurement instruments of which laser emitting extension lines mutually form a 90-degree included angle. By coordinated cooperation of the triaxial sliding mechanism and the laser measurement instruments, the system provided by the invention can complete measurement of side face and upper surface planeness of the measured workpiece in one work procedure, and thus, measurement efficiency is improved.

The invention relates to a laser reception principle based electronic height instrument. The laser reception principle based electronic height instrument has much limitation on its application due to the defects that the instrument is easily disturbed externally, inaccurate and the like. The laser reception principle based electronic height instrument comprises laser emission heads, laser reception heads, a main circuit board and support rods. The support rods are used for height adjustment of the height instrument. The laser emission heads are connected with a driving board. The laser reception heads are connected with the main circuit board. According to the height instrument, touch height is detected through vertical laser arrays; when lasers at a certain height are blocked off, the laser reception heads at the corresponding height do not receive correlation laser signals, and thus, the highest touch height of a testee can be judged. By the laser reception principle based electronic height instrument, the problems of low sensitiveness, low accuracy and incapability of moving of existing height instruments are solved effectively.

The invention relates to a medical image processing method and device, a computer device and a readable storage medium. The method comprises the steps of obtaining a lying morphological structure image of a detection object; inputting the lying morphological structure image into a generative adversarial network model to obtain a prediction image; wherein the prediction image is a standing morphological structure image of the detection object under the magnetic field intensity corresponding to the lying morphological structure image. The method solves the problem that in the prior art, a standing type magnetic resonance image obtained through a low-magnetic-field detection mode is not accurate enough.

The invention discloses a target and background infrared scene simulation method. The method comprises the steps of obtaining infrared features of a target in a target and background infrared scene; And adjusting the infrared image features of the target and the background in the target and background infrared scene according to the infrared features of the target.

The invention discloses a floating and sinking controllable oxygenation device and a control method, and belongs to the technical field of water oxygenation equipment. The device comprises an aeration pump and an aeration outer pipe, the aeration pump is arranged on the upper portion of a first floating block, the aeration outer pipe is of a cylindrical structure, the upper end of the aeration outer pipe is connected with the bottom of the first floating block, and an impeller is arranged at an outlet in the bottom end of the aeration outer pipe; an aeration connecting pipe body which penetrates through the first floating block and is connected with an air outlet of the aeration pump and an auxiliary connecting ring body are arranged in the aeration outer pipe, the auxiliary connecting ring body is arranged below the impeller, a spacing distance is formed between the auxiliary connecting ring body and the impeller, and the inner ring diameter of the auxiliary connecting ring body is larger than the outer diameter of the aeration outer pipe; the auxiliary connecting ring body is connected with the outer wall of the aeration outer pipe through an auxiliary supporting frame. According to the device, the problems that traditional aeration equipment, an impeller type aeration device and an aeration device are not high in aeration efficiency and not accurate enough and waste resources are solved.

The invention discloses a spark plug motor gap accurate adjusting device which comprises a box body, an arrangement space is arranged in the box body, an adjustment mechanism is arranged in the lowerside of the arrangement space, the adjustment mechanism comprises a first sliding groove formed in the inner wall of the lower side of the arrangement space, a first sliding block is arranged in the first sliding groove in a sliding mode, and a first electromagnet is fixedly arranged in the right end surface of the first sliding block. According to the invention, spark plug fixed-gap fine adjustment work can be automatically completed, a spark plug is controlled by the motor to rotate so that the spark plug can be screwed into the threaded hole and move up and down in the threaded hole, and inthe rotating process, the spark plug side electrode is clamped into the groove; the clamping groove is right opposite to the sliding groove, so that the sliding block drives a detector to enter sideelectrode and center electrode spaces, and the sparking plug rotates at a high speed and moves to the detection mechanism through two pairs of gears with different transmission ratios.

The invention discloses a living body identification method, a storage medium and terminal equipment. The living body identification method comprises the steps of: providing to-be-identified image data which comprises continuous N frames of to-be-identified images, wherein N is a positive integer greater than 1; acquiring a chrominance signal corresponding to each frame of to-be-identified image by means of a RGB signal of the image; carrying out Fourier transform on the chrominance signals to obtain N frequency domain sequences, performing recurrent neural network analysis on the N frequencydomain sequences through using a network, and acquiring heartbeat data through iteration; judging whether the heartbeat data exceeds a preset heartbeat threshold range or not, and if so, judging thatthe to-be-identified image data does not contain a living body; and when judging that the heartbeat data does not exceed the preset heartbeat threshold range, judging whether moire patterns exist in the to-be-identified image data or not, and if so, judging that the to-be-identified image data does not contain a living body, otherwise, judging that the to-be-identified image data contains a livingbody. The living body identification method solves a problem that a living body identification method in the prior art is not accurate enough.

The invention relates to the field of intelligent decision, and discloses an environmental pollution grading method, which comprises the following steps: iteratively adjusting an original neural network according to waveform sampling data and corresponding early warning value domain data to obtain a target neural network; collecting original concentration waveforms of various pollutants; randomly constructing a fitting concentration waveform, and establishing a target function by using the difference between the fitting concentration waveform and the original concentration waveform; optimizing the fitting concentration waveform by using a constructed iterative formula and the original concentration waveform to obtain a target concentration waveform; and performing pollution concentration grading treatment on the waveform parameters of the target concentration waveform by using the target neural network to obtain pollution grading results of the various pollutants. The invention further provides an environmental pollution grading device, equipment and a storage medium. The invention also relates to a block chain technology, and the waveform sampling data can be stored in a block chain node. According to the invention, the problems of low pollutant detection efficiency and insufficient accuracy can be solved.

The invention relates to the field of highway engineering, in particular to pavement thickness detection equipment for highway engineering. The equipment comprises a supporting base, wherein universal wheels are fixedly connected to the bottom of the supporting base, a driving motor is fixedly connected to the inner wall of the supporting base, a driving worm is fixedly connected to an output shaft of the driving motor through a coupler, and the top of the supporting base is movably connected with a first connecting shaft. According to the pavement thickness detection equipment for the highway engineering, the universal wheels are arranged, so that the supporting base can be conveniently moved, and the detection position can be conveniently adjusted. Meanwhile, the thickness of different positions of a pavement can be conveniently and synchronously detected through a detection dial indicator in the moving process, so that the accuracy of thickness detection is improved, the problem that the pavement thickness detection is not accurate enough is solved, the labor consumption of detection work is reduced at the same time, the operation convenience is improved, the working requirements of people are met, and the equipment is worthy of popularization and application.

The invention discloses a liquid medicine testing device and a testing method based on the same. The testing device comprises a machine body; a bottle rack is arranged at the top of the machine body;a processor and a wiring board are arranged in the machine body; an LED indicating lamp is arranged on the front surface of the machine body; a nameplate is arranged on the back surface of the machinebody; USB interfaces, a controller interface, an MSB interface and a power junction box are sequentially arranged on one side of the back surface of the machine body from top to bottom; a measuring interface is arranged on the other side of the back surface of the machine body; a bottle fixing clamp is arranged on the bottle rack and is used for fixing two reagent bottles; the nameplate comprisesdescriptions related to a power voltage, an equipment type and a product serial number; and the two USB interfaces adopt A-type USB ports and are connected with equipment such as a printer and the like. The working efficiency of a laboratory can be better improved, and the accuracy of a liquid medicine testing result can be better realized; an automatic liquid medicine testing machine is adoptedto improve the working efficiency and realize better humanized management; and the problems that an existing testing technology is insufficiently accurate in testing result and low in testing efficiency are solved.