58results about How to "The amount of computing data is small" patented technology

The invention discloses an unsupervised image division method based on a multi-target immune cluster integration technology, which mainly solves the problems of poor global optimization capability, single evaluation index, single division scheme form and difficult selection of a plurality of division schemes in the traditional technology. The method comprises the following steps of: (1) extracting gray scale information and wavelet energy information of an image to be divided; (2) sampling the image by using an area-based sampling policy to generate a test sample set; (3) selecting different characteristic vectors to form a plurality of test sample sets; (3) generating a primary division scheme by using an evolution cluster based on a multi-target immune algorithm; (5) integrating and learning an optimal division scheme in the primary division scheme; (6) marking the class of the pixel points of the image according to the selected division scheme; and (7) outputting the image division result. The invention has advantages of high average accuracy in image division and strong robustness and is applicable to the obtaining of image information and the division of image texture.

The invention relates to an infrared array based people stream behavior analysis system and method. The analysis system is characterized by comprising a detection terminal, a transmission module and a processing background, the detection terminal transmits a temperature data packet to the transmission module, and the transmission module transmits the temperature data packet to the processing background. The analysis method is characterized by comprising a step of human identification and people stream counting, a step of calculating stay time of the people stream and a step of calculating the people stream. The system and method have the advantages that people indoor can be counted conveniently in real time, interference is eliminated effectively, and the statistical error is reduced.

A guide information trigger method based on a scenery-viewing area in a global positioning system (GPS) intelligent guide system mainly comprises the steps of collecting scenery-viewing area related data and triggering guide information broadcast by GPS locating information processing. The triggering step comprises obtaining user real-time location information through a GPS; extracting location attributes of a to-be-reached area in a scenery-viewing area database; judging a user location and attribute data of a to-be-reached scenery-viewing area meet the conditions; judging whether scenery-viewing area guide information corresponding to the scenery-viewing area meets trigger conditions; and broadcasting the guide information according to broadcasting rules. The scenery-viewing area is rectangular, is easy to distribute and is reasonably arranged, and data collection quantity and workload are small. A judging method based on the to-be-reached scenery-viewing area is efficient and easy to use. The provided guide information can meet users' scenery-viewing requirements, missing of the guide information caused by system judgment is avoided, and safety prompt information also helps users to improve safety awareness and obtain good visiting experience.

The invention discloses a method for embedding a robust and reversible watermark. The method includes: watermark embedding and watermark extraction. The method has the advantages that the features of a robust watermark and a reversible watermark are fully utilized, the robust watermark and the reversible watermark are combined to provide the robust and reversible watermark which can resist JPEG compression attacks, noisy points of images are filtered through low-pass filtering, robustness is solidified, image features are identified through the edge detecting technology, image partitioning is performed, the to-be-embedded block of the robust image is selected by the image texture pixel, obtained by edge detection, of each block, so that the position of embedding selected block after jpeg compression is consistent with the position of the block selected and extracted by the same manner; the watermark is embedded to increase the robustness of the watermark, and the JPEG compression attacks are resisted.

The invention provides a micro-fluidic chip, an analysis device and a control method of the micro-fluidic chip. The micro-fluidic chip comprises multiple detection electrode assemblies arranged at intervals in the flowing direction of micro-droplets, wherein each detection electrode assembly comprises a first detection electrode and a second detection electrode which are spaced mutually and are used for forming a loop with the micro-droplets so as to determine the position of the micro-droplet. The micro-fluidic chip can realize real-time positioning of the micro-droplets, is low in cost and facilitates integration and miniaturization, interference of a biochemical reaction of the environment and the micro-droplets can be avoided through feedback by current, the operation data volume can be reduced only on the basis of existence of the micro-droplets, the reliability is increased, and driving of the micro-droplets cannot be affected due to low voltage.

The invention discloses a lane line generation method and device, a computer readable storage medium and a terminal, and the method comprises the steps: carrying out the plane fitting of a part of scene point cloud data, so as to obtain a pavement function; extracting points from the scene point cloud data, wherein the distance between the points and a plane represented by the pavement function is smaller than or equal to a preset distance, so as to obtain pavement point cloud data; dividing at least one part of the road surface point cloud data into a plurality of voxels; the method comprises the steps of sequentially determining a height difference value of two adjacent voxels on a plurality of planes perpendicular to a forward direction according to the forward direction of a vehicle and a preset sequence, and taking at least a part of points in the voxels with relatively large height values as boundary feature points when the height difference value is greater than a preset threshold value; and generating the lane line according to the boundary feature points. In the vehicle driving process, the boundary of the lane can be accurately detected in real time, and the accurate and stable lane line is automatically generated.

The embodiment of the invention relates to the field of manicuring, in particular to a manicuring device nail area image acquisition method and system, a manicuring device and a medium. The method comprises: obtaining a user nail image; Obtaining a first color and a second color from the user nail image; Obtaining the contrast ratio of the first color and the second color; And if the contrast is greater than a preset contrast threshold, starting a preset nail segmentation algorithm to obtain a nail region image. According to the technical scheme provided by the embodiment of the invention, when the nail beautifying device does not have to place fingers of a user or place other paperboards, or when the difference between the color of the nail primer coated on the nail of the user and the color of the surrounding environment is small, the preset nail segmentation algorithm is not started, the operation data volume is reduced, and the effect of prolonging the service life of the processoris achieved.

The invention relates to the field of powder cake production equipment, and discloses a full-automatic powder pressing machine for powder cake production. According to the full-automatic powder pressing machine for powder cake production, on one hand, a full-cycle powder cake automatic production process can be realized, wherein the full-cycle powder cake automatic production process comprises automatic filling of packing trays, automatic movement of powder pressing molds, automatic mold closing and pressing of powder cakes and automatic demolding, and therefore, manual workload and working intensity can be greatly reduced, and the adverse influence of dust on physical health of workers can be avoided; on the other hand, a visual system and an image processing technology do not need to beintroduced, and therefore, not only can the operation data amount be greatly reduced, but also the automatic speed can be accelerated, furthermore, mechanical faults caused by the dust can be avoided, long-term maintenance is facilitated, and the service life of a machine is prolonged; and meanwhile, separation of a female mold and a male mold in the powder pressing molds can be realized, and through the backward move of the female mold and the upward move of the female mold when mold closing and pressing are in place, the powder feeding action, the mold closing and pressing action and the demolding action can be automatically realized, and compactness and completeness are achieved.

The invention discloses a visual inspection and obstacle avoidance system for an industrial robot. The visual inspection and obstacle avoidance system comprises an industrial robot body, a camera module, a control module, a block-chain module and a wheel train module, wherein the camera module is used to acquire a real-time image of a front end of the industrial robot; the control module is used for receiving a target instruction sent from a user terminal to perform an obstacle avoidance task; the block-chain module is used for encrypting state information of the industrial robot and the target instruction and uploading the encrypted state information and target instruction to a blockchain; and the wheel train module is used for controlling the direction and turning angle of a plurality ofdriving wheels. The visual inspection and obstacle avoidance system for the industrial robot can plan an obstacle avoidance path more reasonably, is lower in failure rate, can handle a possible faultin time, and is suitable for the industrial robot operating at a high speed.

The invention discloses an effective vibration waveform intercepting method based on three-level threshold judgment, and the method comprises the steps: vibration data is acquired, an amplitude average of the vibration data is calculated, and if a signal is caused by large-scale environmental disturbance is judged through comparing the amplitude average of the vibration data and a global threshold; if the signal is a vibration signal is judged through comparing an amplitude of the vibration data and an amplitude threshold T1 of invaded data, and waveform data in certain length before and after the vibration data of the vibration signal is intercepted; fragment division is performed on the intercepted waveform data, vibration energy in each fragment is calculated and smoothed, and if the vibration waveform is effective is judged through comparing the number and threshold of fragment of which vibration energy serially exceeds a fragment vibration energy threshold. According to the invention, a three-level threshold is adopted to judge, misinformation caused by environment is prevented, calculation data volume is reduced, processing time is effectively shortened, and positioning precision of an optical fiber perimeter intrusion monitoring instrument.

The invention relates to the medical field, in particular to a monitoring system for preventing children from falling. The monitoring system comprises an LED warning device, a brightness enhancing device, an intelligent imaging structure, and the like. The LED warning device is used for flicking at a preset frequency to give a corresponding field alarm when receiving a field alarm signal. The brightness enhancing device is used for receiving an internal environment image and enhancing the brightness of the internal environment image to obtain and output a brightness enhanced image. By using the monitoring system of the invention, the protection ability can be improved, and children can be prevented from falling accidentally due to hyperactivity or playfulness in the absence of doctors.

The invention provides a furnishing theft-proof detection device and a detecting method thereof. In the furnishing theft-proof detection device, a local furnishing surface or an object surface adhered on the furnishing surface is taken as a detecting area, the characteristic that furnishings are usually placed in fixed points and fixed places is utilized, and an alarm signal is sent out when displacement of the furnishings is detected, so that interference of environmental factors is greatly reduced, the theft-proof detecting sensitivity of the furnishings is improved, the structure is simple, and the cost is low; and the device has excellent industrial production and market popularization prospects. The detecting method for the furnishing theft-proof detection device uses grey values of pixels in an imaging image in the detecting area as a detection judgment basis, has higher space sensitivity and time sensitivity and a simple algorithm, can be easily realized through a programming technology by common programming technicians without a special software technical platform, and has a wide application range.

The invention discloses a correlation mining method and device for a large user abnormal power consumption analysis data set, and the method comprises the steps of carrying out the discrimination of the support degree of the large user abnormal power consumption analysis data set, and obtaining a plurality of screened frequent item sets; and carrying out confidence judgment on the abnormal power utilization analysis data set through the plurality of screened frequent item sets to obtain a large user abnormal power consumption analysis data set with correlation. According to the invention, thecorrelation between the item sets of the abnormal power consumption analysis data set of the large user is mined through judgment of the support degree and the confidence degree, the operation data volume of the abnormal power utilization analysis algorithm is reduced, the system analysis efficiency is effectively improved, and the invention has a good application prospect.

The embodiment of the invention discloses a self-adaptive order aggregation method and electronic equipment, and the method comprises the steps: determining an order aggregation limiting parameter ofthe electronic equipment based on the order data of the electronic equipment; receiving screening parameters for the order data, and screening the order data based on the screening parameters to obtain screened order data; aggregating the screened order data based on the aggregation limiting parameters to obtain aggregated order data; wherein the aggregated order data and the order data containedin the aggregated order data meet the conditions limited by the aggregation limiting parameters; recording a corresponding relationship between the aggregated order data and the order data contained in the aggregated order data; and determining components required by the electronic equipment based on the aggregated order data, and collecting the components to assemble the electronic equipment matched with the electronic equipment in the aggregated order data in number.

The invention discloses a wire diameter anomaly identification method and device, electronic equipment and a computer readable storage medium. The method comprises the steps of preprocessing a currentbackground image to obtain a background image of a target wire and a partition where the target wire is located; performing threshold segmentation to obtain a first image, and recording the number ofeach partition; performing connected domain filtering on the first image; performing interval device interference removal and similar object interference removal on the first image to obtain a secondimage; in the second image, taking a point of which the gray value is greater than the gray value of the black wire in the target wire as a seed point, and then performing regional growth by taking the seed point as a starting point to obtain a black connected domain of each wire; and extracting the edge of the black connected domain, performing wire edge checking calculation on the two oppositelong edges, performing linearity checking calculation judgment, and when the linearity of the two long edges is greater than a standard threshold, determining that the wire diameter is abnormal. According to the invention, targeted and efficient diameter anomaly analysis and identification processing can be carried out.

The invention relates to a steering wheel real-time positioning platform which comprises a titanium dioxide type oxygen sensor arranged in an exhaust pipe of a vehicle and used for detecting the oxygen concentration in the exhaust pipe; the titanium dioxide type oxygen sensor comprising a titanium dioxide element which is a circular plate type electrode, and a platinum wire being placed in the middle of the titanium dioxide element; an explosion vibration sensor connected with the engine cylinder body and used for detecting the explosion vibration state of the engine cylinder body; wherein theexplosion vibration sensor comprises a shell, a permanent magnet, a strong magnet iron core and a coil, the strong magnet iron core being excited by the permanent magnet, and the coil being wound around the strong magnet iron core; and a distance measuring device connected with the target analysis equipment and used for determining the distance from the centroid of the steering wheel pattern to the centroid of the mobile terminal pattern in the second adjustment image so as to output the distance as instant measurement distance. According to the platform, the driving violation behavior can beeffectively detected.

The invention relates to an infrared array-based induced abortion behavior analysis system and human body identification method. The analysis system is characterized by comprising a detection terminal, a transmission module and a processing backstage, wherein the detection terminal is used for transmitting temperature data to the transmission module, and the transmission module is used for transmitting the temperature data to the processing backstage. A data analysis method based on the analysis system is also provided. The beneficial effects of the infrared array-based induced abortion behavior analysis system and the human body identification method are that human body identification can be reliably realized, and big data statistical operation on the number of people and big data analysis of induced abortion behavior can be conveniently performed.

The invention is suitable for the technical field of radar, and provides a dense multi-target parameter extraction method and device and terminal equipment, and the method comprises the steps: readinga pre-processed to-be-detected image in any distance door; calculating a gray scale gradient maximum value in the to-be-detected image, determining edges of a plurality of targets in the to-be-detected image, binarizing on the to-be-detected image of which the edges of the plurality of targets are determined, and determining a binary image matrix corresponding to the to-be-detected image; determining minimum bounding rectangle data corresponding to each target area according to the binary image matrix; calculating the pixel point with the maximum amplitude in each minimum bounding rectangle data, and taking the pixel point with the maximum amplitude as a mass point corresponding to each target; and determining the time, speed and amplitude data of the target according to each mass point.Therefore, the time, the speed and the amplitude of each target can be effectively extracted, and conversion from digital signals to physical parameters is realized.

The invention discloses a method for segmenting bleeding areas in brain CT images based on semi-supervised learning. The method includes a semi-supervised model training stage and a semi-supervised model-based bleeding area segmentation stage; the semi-supervised model training stage is used to train semi-supervised model; the bleeding area segmentation stage based on the semi-supervised model includes format conversion of the two-dimensional CT image sequence that needs to be segmented into the intracranial hemorrhage area, reconstructing the two-dimensional CT image into a three-dimensional space, and then using the supervoxel algorithm to divide the three-dimensional image into For supervoxels of similar size, features are extracted from each supervoxel as a sample, and finally the supervoxels are divided into foreground and background parts through a trained semi-supervised model based on the features. The present invention effectively improves the accuracy of bleeding area detection by introducing a semi-supervised learning algorithm and using supervoxels instead of pixels for calculation.

The invention discloses an unsupervised image division method based on a multi-target immune cluster integration technology, which mainly solves the problems of poor global optimization capability, single evaluation index, single division scheme form and difficult selection of a plurality of division schemes in the traditional technology. The method comprises the following steps of: (1) extracting gray scale information and wavelet energy information of an image to be divided; (2) sampling the image by using an area-based sampling policy to generate a test sample set; (3) selecting different characteristic vectors to form a plurality of test sample sets; (3) generating a primary division scheme by using an evolution cluster based on a multi-target immune algorithm; (5) integrating and learning an optimal division scheme in the primary division scheme; (6) marking the class of the pixel points of the image according to the selected division scheme; and (7) outputting the image divisionresult. The invention has advantages of high average accuracy in image division and strong robustness and is applicable to the obtaining of image information and the division of image texture.

The invention relates to a dust particle density analysis system which comprises a falling sensing device arranged in a dust removal robot and used for sending out a first control signal when sensingthat the dust removal robot falls, otherwise, sending out a second control signal; The soft cushion release equipment which is arranged below the dust removal robot, is connected with the falling induction equipment and is used for releasing a soft cushion in the dust removal robot when receiving the first control signal and recovering the released soft cushion when receiving the second control signal; The motor driving equipment which is used for determining a corresponding motor rotating speed based on the received dust density so as to drive a dust suction motor of the dust removal robot tooperate according to the motor rotating speed; And the density analysis equipment which is used for identifying each dust particle in the image based on the dust particle image characteristics and determining the corresponding dust density based on the number of the dust particles in the image. According to the invention, the dust removal precision of the dust removal robot can be improved.

The invention belongs to the field of holography and three-dimensional display, and particularly relates to a computer-generated holography three-dimensional display method and system. A human eye observation position is traced and identified by a human eye tracking sensor, three-dimensional object information received within the human eye observation position is only reproduced, the strength, theposition and the propagation direction of a voxel point are accurately controlled by a mode that a reproduced voxel point of a hologram is loaded by a spatial light modulator, a gray value of the hologram corresponding to shielded information of each voxel point is set, so that a shielding relation of the voxel point is correctly expressed, dynamic three-dimensional display can be achieved by combining the spatial light modulator to refresh and load the computer-generated hologram, the method is simultaneously and suitably used for three-dimensional real image display and three-dimensional virtual image display, the computer-generated holography dynamic three-dimensional display capable of reducing reproduced waste information can be achieved, the operation data volume is reduced, and theshielding relation of a reproduced image of a three-dimensional object can be correctly expressed.

The embodiment of the invention provides an encoding mode selection method and device, electronic equipment and a computer readable medium. The method comprises the following steps: respectively estimating a symmetrical inter-frame prediction mode and a motion merging prediction mode for a prediction unit to obtain rate distortion costs of the symmetrical inter-frame prediction mode and the motionmerging prediction mode; determining a first candidate prediction mode according to the rate-distortion cost of the symmetric inter-frame prediction mode and the rate-distortion cost of the motion merging prediction mode; determining a second candidate prediction mode set from the asymmetric inter-frame prediction modes according to the first candidate prediction mode; carrying out candidate asymmetric inter-frame prediction mode estimation to obtain the rate distortion cost of a second candidate prediction mode set; and determining a target prediction mode of the target coding unit accordingto the rate-distortion costs of the first candidate prediction mode and the second candidate prediction mode set. According to the technical scheme, the second candidate prediction mode set is determined under the guidance of the first candidate prediction mode, the operation data volume can be effectively reduced, and the encoding speed is increased.

The invention provides a method for detecting pedestrians, which adopts an obliquely installed fisheye camera to collect images, and ensures that pedestrians can be detected at an optimal monitoring angle without fisheye distortion correction. The obtained sample set is used for training, the aggregated channel features are extracted by the ACF method, and the classifier is trained by the soft cascade AdaBoost method, which can improve the accuracy of pedestrian statistical features without fisheye distortion correction; in the detection In the process, after quickly extracting 8 features, use a fixed-size pyramid aggregation template to aggregate 8 feature images to form pyramid aggregation channel features, which can avoid using all pyramid channel features, thereby reducing the amount of computing data. Improve the calculation speed, and finally merge the overlapping candidate windows through the greedy non-maximum value suppression method to improve the robustness and accuracy of pedestrian detection.

The invention provides an image segmentation method based on a similarity interaction mechanism, which mainly solves the problems of low segmentation efficiency, poor region homogeneity and detail information loss of the prior segmentation algorithm. The implementation steps of the image segmentation method are as follows: (1) the characteristics of each pixel point of an image to be segmented are extracted; (2) the characteristics of region blocks are obtained; (3) the similarity among the region blocks is calculated; (4) phase values of the region blocks are obtained; (5) the region blocks are classified; and (6) an image segmentation result is output. The image segmentation method has the advantages of high segmentation efficiency, strong region homogeneity of the segmentation result, much detail information and good edge effect, can be effectively used for segmenting texture images and synthetic aperture radar (SAR) images and can be used for recognizing image targets.

The invention provides a microfluidic chip, an analysis device, and a control method for the microfluidic chip. The microfluidic chip includes: a plurality of detection electrode groups, and the plurality of detection electrode groups are arranged at intervals along the direction of micro-droplet flow, each Each of the detection electrode groups includes a first detection electrode and a second detection electrode spaced apart from each other, and the first detection electrode and the second detection electrode are used to form a circuit with the micro-droplet to determine the micro-droplet s position. The microfluidic chip can realize the real-time positioning of micro-droplets, and the cost is low, which is conducive to integration and miniaturization. The use of current feedback can avoid the interference of the environment and the biochemical reaction of micro-droplets. No basis can reduce the amount of calculation data, increase reliability, and the applied voltage is low, which will not affect the driving of micro-droplets.

The invention relates to the technical field of drop point measurement, and provides a target drop point measurement method based on building outline features, which uses an RTK inker to measure building outline point coordinates and predicted drop point coordinates; Immediately stop the recording after the fire appears, and calculate the number of backward images n according to the movement time of the target in the camera's field of view and the frame rate of the camera; screen the target detection range by the frame difference method; combine the Otsu threshold segmentation algorithm to process the differential image , manually screen in sequence, select the key frame image of the moment when the target passes through the roof, and click on the image to select the position of the building outline point and the actual landing point of the target; The coordinates of the landing point; use the coordinate axis rotation formula to obtain the target landing point data in the shooting coordinate system. The present invention only needs to measure part of the existing contour points before the test, and does not need to carry out the work of marking point layout, which effectively reduces the difficulty of field construction.

The invention discloses a satellite information operational method for a global positioning system device. The method comprises the following steps of: providing a antenna module for receiving satellite signals; providing a signal processing circuit for performing analog-to-digital signal processing on the satellite signals and acquiring a plurality of packets of first digital data; providing a memory which stores a check list, wherein the check list records corresponding relations between a plurality of packets of second digital data and a plurality of packets of return values; and providing a central processing unit for grouping each of the first digital data in turn by utilizing number of any packet of the second digital data in the check list so as to obtain each of the second digital data corresponding to each of the grouped first digital data through comparison, and searching out each corresponding return value according to each obtained second digital data through comparison to calculate continuously according to each searched return value.

The invention discloses a method for robust and reversible watermark embedding, including: watermark embedding and watermark extraction; fully utilizing the characteristics of robust watermark and reversible watermark, and combining the two, a robust reversible watermark which can resist JPEG compression attack is proposed , through low-pass filtering, the noise on the picture is filtered out, and the robustness is consolidated. The edge detection technology identifies the picture features, and then divides the picture into blocks. To select the block to be embedded in the robust watermark, after the selected block is compressed by jpeg, when the extracted block is selected in the same way, it is consistent with the position of the previously embedded selected block; the watermark is embedded to increase the robustness of the watermark, Resists JPEG compression attacks.