46results about How to "Reduce the number of acquisitions" patented technology

The invention relates to a correction method of an LED display screen. The correction method comprises: sizes of modules for correction and a separated module collection mode of an LED display screen are determined; the LED display screen is controlled to be lightened and primary-color image collection is carried out, wherein each lightened correction module contains a plurality of controlled lightened same-color LED lamp points with continuous positions; according to the collected primary-color image, an initial brightness module or an initial brightness chromatic value of each lightened correction module is obtained and a brightness correction target value or a brightness chromatic correction target value is determined correspondingly; and on the basis of the initial brightness module or an initial brightness chromatic value of each lightened correction module as well as the corresponding brightness correction target value or a brightness chromatic correction target value, brightness correction coefficients or brightness chromatic correction coefficients of the multiple same-color LED lamp points in each correction module are obtained and are applied to the LED display screen. According to the invention, the number of LED lamp points for single-time collection is increased by using a separated module collection way and the number of times of image collection is also reduced; the correction time can be shortened; and the correction efficiency can be improved.

The invention belongs to the field of image identification, and particularly relates to a stick counting method and a stick counting device. The stick counting device comprises an image acquisition module used for acquiring an end face image of sticks, an image processing module used for processing the acquired end face image of sticks, an image identification module used for identifying the number of sticks based on the processed end face image of sticks, and a result output module used for outputting the identified number of sticks. The problem that manual counting is time-consuming and effort-consuming and causes a big chance of error is solved. The number of sticks can be obtained immediately with high accuracy.

The invention discloses a terminal device and a fingerprint identification method and device, and belongs to the field of a smart terminal. The terminal device comprises a processing chip, a fingerprint sensor and a temperature collection assembly. The processing chip is electrically connected with the fingerprint sensor and the temperature collection assembly. The processing chip is used for collecting a fingerprint signal through the fingerprint sensor and collecting a temperature signal through the temperature collection assembly, determining a compensation coefficient corresponding to the temperature signal, compensating the fingerprint signal according to the compensation coefficient and carrying out fingerprint identification according to the compensated fingerprint signal. The problem that the whole fingerprint identification process is relatively long due to the fact that in order to avoid an identification error of the fingerprint signal resulting from the environment temperature, the processing chip needs to adjust hardware parameters and processing algorithms continuously and collect the fingerprint signal for many times for identification is solved, the collection times of the fingerprint signal is reduced, the fingerprint identification process is simplified, and the fingerprint identification speed is accelerated.

Embodiments of the invention disclose a magnetic resonance imaging method and system. The method comprises the following steps: applying non-layering inverse pulses to a detected object; simultaneously stimulating multiple layers of a heart by using a preset imaging sequence in a set heart movement phase in a first heartbeat period, and acquiring a multilayer aliasing T1 weighted image of the heart; acquiring a reference image by using a preset imaging sequence stimulated by a preset overturning angle in a set heart movement phase in a second heartbeat period, wherein the first heartbeat period and the second heartbeat period correspond to a same breath-holding and are two adjacent heartbeat periods, and the reference image corresponds to one layer of the multiple layers; and performing the interlayer de-aliasing and phase-sensitive reverse restoration re-establishment for the multilayer aliasing T1 weighted image by using the reference image to obtain a target T1 weighted image of each layer of the heart. By virtue of the technical scheme, on the premise of ensuring the image quality, the magnetic resonance data acquisition efficiency can be improved.

The invention relates to the technical field of door lock security control, and in particular discloses an intelligent fingerprint door lock and a fingerprint record method thereof. The intelligent fingerprint door lock comprises a lock body and a mobile terminal, wherein the lock body is in wireless communication connection with a mobile terminal; the lock body comprises a fingerprint processing device and a mechanical lock; the fingerprint processing device is electrically connected with the mechanical lock. According to the intelligent fingerprint door lock provided by the invention, in the process that fingerprint information is acquired by using a fingerprint acquisition module of the fingerprint processing device, wrong fingerprint data are corrected, the fingerprint acquisition quality of users is improved, and the false rejection rate of fingerprint unlocking is reduced; meanwhile the fingerprint acquisition times are calculated by using a counter, by controlling the fingerprint acquisition times, and together with fingerprint acquisition posture instructions of the mobile terminal, fingerprint acquisition accuracy is further effectively controlled, and the false rejection rate of fingerprint unlocking is reduced.

The embodiment of the invention provides a pointer type instrument identification method and electronic equipment, and relates to the field of image processing. A processor processes a pointer type instrument image to obtain an intersection point of a pointer and a dial in the pointer type instrument image, and the reading of a pointer type instrument is determined through the position of the intersection point on the dial; compared with a manual reading mode in the prior art, the manual cost is reduced, and the reading efficiency of the pointer instrument is improved. Compared with the priorart that a large number of sample images of the pointer type instrument are needed when the pointer type instrument image is read through the neural network, the method can achieve the obtaining of the scale value pointed by the pointer in the pointer type instrument image without a large number of samples, and improves the recognition efficiency.

The invention relates to a maintenance method for tree seedlings, in particular to a crossbreeding method for paper mulberry. The crossbreeding method comprises the following steps: (1) preparing a paper mulberry pollinating device; (2) artificially pollinating a male parent and a female parent, sowing and cultivating after the fruit is mature; (3) artificially pollinating a novel generation of hybrid paper mulberry, sowing and cultivating after the fruit is obtained; and (4) scientifically managing the newly grown seedling. The crossbreeding method solves the problems of allergic reaction of a pollen collecting worker and blockage of a nozzle of a pollinator caused by excessive pollens in the artificial pollinating process.

The invention relates to a yarn monitoring device and a yarn winder. The yarn monitoring device (15) includes a sensor unit (35), an obtaining section (55), a setting section (58) and an evaluation section (53). The sensor unit (35) outputs a detection value corresponding to the presence/absence of the yarn (10) in the detection area (36) and/or a state of the yarn (10). The obtaining section (55) obtains the detection value in a state in which no yarn (10) exists in the detection area (36). The setting section (58) sets an evaluation reference value based on at least one of a plurality of detection values which are obtained by repeating the detection of the detection value by the obtaining section (55) until the yarn (10) is inserted into the detection area (36).The evaluation section (53) evaluates the state of the yarn (10) based on the evaluation reference value set by the setting section (58).

The invention discloses a living body acquisition rolling and flat printing fusion acquisition fingerprint method and relates to the fingerprint identification technology field. In the method, fingerprints of different positions of fingers are collected in steps and finally a complete finger fingerprint is acquired. During an acquisition process, for a person whose fingerprints need to be collected, a difficulty of rolling the fingers is reduced, handleability of fingerprint acquisition is increased, and a problem that collected fingerprint quality can not be guaranteed because finger edges are not sufficient, a force is not uniform, rotation is not coherent and so on is solved. The collected fingerprint quality is greatly increased. In addition, during one acquisition, a flat printing fingerprint and/or a rolling fingerprint can be acquired simultaneously. In the prior art, the flat printing fingerprint and the rolling fingerprint need to be collected for two times. Therefore, in the invention, an acquisition frequency of the flat printing fingerprint and the rolling fingerprint is reduced by half.

The invention discloses a fast color fringe pattern phase demodulation algorithm and system, and the method comprises the steps: firstly carrying out the pre-calibration of a measurement system through a calibration flat plate, and solving a coupling strength coefficient; secondly, using the system to respectively project a uniform intensity graph and a color fringe graph to an object to be measured, after the two graphs are respectively decoupled, using the uniform intensity graph to solve a background item and a modulation degree, and performing background removal and color normalization on the color fringe graph to obtain three background-free and color-normalized fringe patterns; adopting a phase shift method to solve demodulation phases of the three fringe patterns; and finally, unwrapping the obtained demodulation phase to obtain an unwrapped phase. The phase shift method demodulation of the single color fringe pattern is realized after background item calibration, the robustness of the demodulation of the traditional gray phase shift method is reserved, the fast phase demodulation is realized, and the requirement of dynamic three-dimensional shape measurement is met.

The invention relates to an idle conference room detection method based on an integral algorithm, and the method specifically comprises the steps: S1, obtaining a reservation request, and generating a conference room reservation table; S2, detecting whether the conference room is idle or not, if not, turning to S3, and if yes, ending; S3, judging whether the conference room is a video conference or not, if so, turning to S4, and otherwise, turning to S5; S4, judging whether video connection exists or not, if not, turning to S5, and if yes, obtaining conference room images shot by a camera in the conference room at regular time, and turning to S6; S5, acquiring conference room audio recorded by a loudspeaker in the conference room at regular time, and turning to S6; S6, setting the basic integral as 0 point, detecting whether the image and the audio of the conference room change in continuous time, if so, increasing the integral by 20 points, otherwise, keeping the integral of the conference room unchanged, judging whether the integral reaches 60 points, and if so, adjusting the use condition of the corresponding conference room to be idle. Compared with the prior art, the method has the advantages of improving the utilization rate of the conference room, and being low in cost and development amount.

The invention discloses an anomaly detection method of a sudden crowd based on instantaneous energy, the method of extracting motion feature points based on pixel statistical analysis is used to extract the feature points of the image motion region, and the kinetic energy is used as the basic energy feature of the crowd, and the instantaneous energy feature of the crowd motion state is extracted based on the kinetic energy difference between two consecutive frames of the energy block, so as to detect the human sudden crowd anomaly. The anomaly detection method of the sudden crowd can not onlyeffectively reduce the number of obtaining motion feature points so as to avoid subsequent processing of some meaningless feature points, but also can reflect the texture feature of the foreground motion region and the distribution of the crowd density, which makes the extracted motion feature points more representative and uniform. Whether in the abnormal occurrence of the response speed or the description of the movement of the crowd state changes, the above-mentioned outburst crowd anomaly detection methods have shown outstanding advantages.