800results about How to "Easy splicing" patented technology

The present invention provides systems and methods that provide images of an environment to the viewpoint of a display. The systems and methods define a mapping surface at a distance from the image source and display that approximates the environment within the field of view of the image source. The system methods define a model that relates the different geometries of the image source, display, and mapping surface to each other. Using the model and the mapping surface, the systems and methods tile images from the image source, correlate the images to the display, and display the images. In instants where two image sources have overlapping fields of view on the mapping surface, the systems and methods overlap and stitch the images to form a mosaic image. If two overlapping image sources each have images with unique characteristics, the systems and methods fuse the images into a composite image.

A method for constructing a composite image comprises determining in an alignment between at least a first and a second image based on the content of the first and second image frames, and combining in a transform domain at least the first and second images with the determined alignment to form a composite image.

The invented product is a fleecebacked laminate that is suitable as a single-ply fleecebacked roofing membrane. The laminate consists of the following: 1) a thermoplastic reinforced planar sheet, 2) a needle punched fleece formed on both sides of the thermoplastic reinforced planar sheet, and 3) a thermoplastic polymeric layer extruded onto one side of the fleece. The fleece is formed by needle punching fibers through the reinforced planar sheet. The reinforced planar sheet serves as a common supporting base the resulting two-side fleece. The thermoplastic polymeric layer is embedded with one side of the fleece fibers. The thermoplastic polymeric layer serves as the top layer of the roofing membrane, and the second side of exposed fleece serves as a felt. The thermoplastic polymeric layer is preferably comprised of an extruded PVC substrate, compounded to have excellent weather resistance, printability and to be nonflammable. The fleece fibers are preferably composed of polyester. The thermoplastic reinforced planar sheet is preferably a scrim that is extrusion coated with a thermoplastic substrate that is compounded to have excellent weather resistance, and to be nonflammable.

The invention provides a ceramic material with a multi-stage duct structure and a manufacture method thereof. The ceramic material is characterized by being provided with a plurality of macropores with millimeter grade, tens of middle pores with millimeter grade exist on the walls of the macropores, micropores with sizes less than microns exist on the walls of the middle pores. The ceramic material is manufactured by using 10-50 percent by mass of waste and 20-60 percent by mass of natural porous mineral and matching with 10-30 percent by mass of clay through the steps of adding water for mixing, forming, drying, preheating and sintering at a temperature of 1000-1200 DEG C, wherein the waste can be expanded at high temperature or has large ignition lost and easy decomposition and volatilization. The ceramic material has excellent water-absorbing and water-retaining properties, better water permeability, excellent absorption and filtration and light weight, and can be used as a water-absorbing material, a water-retaining material, a water permeating material, a sound insulating material, a heat insulating material, an indoor deodorizing material, an absorbing material, a filtering material and the like to be widely applied to various fields.

The invention relates to a ceiling sheet, in particular to an environmental-friendly lead-free PVC ceiling sheet and a manufacturing method and applications thereof. The environmental-friendly lead-free PVC ceiling sheet is obtained by blending the following components according to parts by weight and then extruding to mould: 50-100 parts of PVC resin SG-5, 15-50 parts of nano calcium carbonate, 3-10 parts of calcium zinc stabilizer, 1-3 parts of rutile titanium dioxide, 1-3 parts of CPE impact modifier, 1.5-5 parts of ACR enhancer, 0.2-0.8 part of stearic acid, 0.3-1.0 part of calcium stearate, and 0.2-0.4 part of PE wax. The PVC ceiling sheet is used as a novel environmental-friendly lead-free material of which the lead content is less than or equal to 200PPM, and is applicable to non-toxic places with high environmental requirements such as hospitals, operating rooms, baby rooms, food plants, and the like. The invention also discloses the manufacturing method of the ceiling sheet and ceiling gusset plates manufactured by the sheet.

A camera calibration system for cameras of a vehicle includes a plurality of cameras disposed at a vehicle and having respective fields of view exterior of the vehicle, with the fields of view of two of the cameras overlapping. While the vehicle is moving, an image processor of a control processes image data captured by the two cameras to determine and track matching features in an overlapping region of the fields of view of the two cameras. The control, responsive to image processing of image data captured by the two cameras, determines motion of matching features and minimizes a cost function of the error in the matching of the features. The camera calibration system uses the determined minimized cost function to refine calibration results of the two cameras using a feedback loop.

The invention discloses a lip language recognition method and device, computer equipment and a storage medium. The lip language recognition method comprises the following steps: carrying out standardization processing on a frame rate of an obtained original video, and separating the obtained standard video to obtain an effective audio stream and an effective video stream; tracking a human face in the effective video stream by using a human face recognition algorithm, extracting a mouth area in the human face, and obtaining a frame lip action video so as to obtain a lip image sequence; segmenting the lip image sequence by adopting a sequence segmentation rule to obtain a segmented image sequence; sequentially inputting the segmented image sequences corresponding to the lip image sequences into a lip image recognition model for recognition, and obtaining a lip image recognition result; inputting the effective audio stream into a speech recognition model to obtain a speech recognition result; and when the similarity between the lip image recognition result and the original video reaches a preset value, taking the lip image recognition result as a lip language recognition result of the original video so as to ensure the accuracy of the lip image recognition result.

The invention discloses a Mongolian named entity recognition method based on neural network and a recognition system thereof, and belongs to the technical field of natural language processing. The recognition system comprises a Glove tool, a bi-directional language model component obtaining module, a language model vector obtaining module, a BLSTM vector obtaining module, an attention mechanism layer, a CRF layer and a final model obtaining module. The invention adopts forward and backward cascade neural language model BLSTM to learn context information from a large amount of unlabeled corpus,and introduces the learned context information into a neural network based on BLSTM and CRF by attention mechanism, thereby improving Mongolian named entity recognition efficiency and solving the problems existing in the prior art.

The invention discloses a soft and weak wall rock tunnel advance reinforced support construction method. When a pipe roof advance support construction is carried out on one segment of the constructed tunnel, the construction method comprises the following steps: (1) support arch and guide pipe construction, wherein a plurality of guide tubes for carrying out guiding on a plurality of pipe roof pipes are arranged on a support arch; each pipe roof pipe comprises a pipe body and a drill bit; each pipe body is formed by splicing a plurality of pipe joints; each drill bit is coaxially arranged at the front end of the pipe body; the pipe joints are spiral steel pipes; the pipe roof pipes are first pipe roof pipes or second pipe roof pipes; and the first pipe roof pipes and the second pipe roof pipes are staggered; (2) drilling machine and slip casting machine installation and emplacement; (3) pipe roof advance support construction, which comprises the following processes: first pipe roof pipe drilling and synchronous grouting construction, final hole grouting construction of the first pipe roof pipes, second pipe roof pipe drilling and synchronous grouting construction and final hole grouting construction of the second pipe roof pipes. The method disclosed by the invention is simple in steps, reasonable in design, simple and convenient to construct and good in support effect; and the soft and weak wall rock tunnel advance reinforced support process can be simply, conveniently and quickly finished.

Endless rope containing primary strands, the primary strands containing laid-up secondary strands, the laid-up secondary strands containing rope yarns, wherein the primary strands have been laid up from 3, 4 or 6 secondary strands, wherein the rope contains a splice in at least every primary strand and wherein the rope preferably has been laid up from 3, 4 or (1+6) primary strands or, alternatively, has been braided from 8 or 12 primary strands.

The invention relates to an intelligent drilling television imaging instrument, which mainly consists of a main controller, an underground probe, an underground cable, a windlass, a hub, a stepper motor, a depth counter and a tripod. The main controller is used for collecting and storing the information such as image splicing, video, depth and azimuth angle; the underground cable is used for supplying power to the probe and transmitting a control order, video, angle and the like; the underground probe is used for shooting a picture and video and recording the underground azimuth angle and transmitting the information to an industrial control computer through the cable; the depth counter is used for recording the underground depth of the underground probe; the stepper motor is used for controlling the probe to ascend and descend at a constant speed. An open-type lens is directly arranged in the water, and no glass cover is arranged in front of the lens, so that a shot image of a well wall is clearer, and the color retention is more authentic; a combined-type light source is adopted, so that the well wall and the underground situation can be directly observed; by adopting the unique stepper motor and the depth counting hub device, the probe can be lowered and lifted at a constant speed, and the images are completely spliced; by introducing electronic rock core library software, an electronic rock core file is established.

The invention discloses an amplification primer for a grouper catostomidae mitochondrial genome complete sequence and a design method thereof, relates to groupers, and particularly relates to 16 pairs of amplification primers mainly used for specifically amplifying most grouper catostomidae mitochondrial genome complete sequences and a design method thereof. The invention provides an amplification primer for a grouper catostomidae mitochondrial genome complete sequence, a design method thereof and application thereof. The amplification primer for the grouper catostomidae mitochondrial genome complete sequence consists of 16 pairs of single-chain oligonucleotides, wherein each pair consists of one light-chain primer and one heavy-chain primer. The amplification primer for the grouper catostomidae mitochondrial genome complete sequence can be used in the aspects of variety authentication, geographical population authentication, germplasm resource evaluation and the like.

The invention discloses a sound selection method for waveform concatenation speech synthesis. The method comprises the following steps of: on the basis of an original audio, carrying out hidden markov based model training so as to obtain an acoustic model set and a corresponding characteristic decision tree; inputting a plurality of training texts and on the basis of the characteristic decision tree, searching to obtain related acoustic models so as to obtain corresponding target voice and target syllables; according to similarity of the target voice and corresponding candidate primitives and likelihood probability of each acoustic parameter of the candidate primitives under a current acoustic model, training to obtain a similarity classifier; inputting a random text to be synthesized, removing the dissimilar candidate primitives on the basis of the similarity classifier, selecting the optimal primitive from the residual candidate primitives by utilizing a concatenation cost minimization rule and carrying out concatenation to obtain synthetic speech. The adoption of the method disclosed by the invention can synthesize speech with higher tone quality.

The present invention belongs to the technical field of video monitoring and provides a frame splicing method and device.The frame splicing method comprises: zooming multiple video signals; caching the zoomed video signals; and splicing frames corresponding to the cached multiple digital video signals according to priority of the digital video signals so as to achieve splicing of multiple frames. Since the multiple videos are zoomed and spliced once, the quality loss in frame zooming is reduced and the splicing is easy and convenient.

The invention discloses an ultra-high brightness LED matrix straight-downward backlight module applied to a large-scale LCD which has a size of more than 32 inchs. In an LED matrix module six LEDs are connected in series into a straight row, and LED are arranged longitudinally and transversally to form a square matrix; the length and the row number of a printed circuit board (PCB) extends all along to both sides of a screen visible area of the LCD in the arrangement and combination mode and correspondingly adjusts according to the size of the screen so that the arrangement area of an LED matrix is consistent with a display area of the LCD; LEDs the surfaces of which are welded and packaged with 5.0*5.0mm or 3.5*2.8mm three-core gold-white high-brightness overhead light are selected, color coordinate values are: X: 0.303 and Y:0.293 and a brightness value is more than 7000 mcd and is graded by 0.1V to ensure the consistency of low-brightness work; and a distance between two LEDs is 15.6-16.5mm, every six LEDs are connected in series into a straight row, and a distance between two straight rows is 15.6-16.5mm. The straight-downward ultra-high backlight module manufactured by applying an LED matrix improves the brightness higher than 1200 LUH, not only satisfies the requirement for visual brightness under outside sunshine, but also solves the problems of overlarge power and overheating, and the LED matrix has brightness with a large controllable range and is energy-saving and environment-friendly.

The invention relates to a method and system for binocular fish-eye image splicing and unfolding correction. The method for binocular fish-eye image splicing and unfolding correction includes the steps: utilizing a binocular fish-eye camera to obtain an original binocular fish-eye image, and extracting the left and right two effective circular imaging areas from the original binocular fish-eye image; extracting the characteristic points of the overlapped part in the two effective areas, and matching the characteristic points in pairs; respectively mapping the two effective area images to obtain two space coordinate unit spherical surface images, performing matrix singular value decomposition on the three dimensional space spherical surface, obtaining the rotation relation, that is, the rotation matrix, between the two images through calculation, and correcting one of the space coordinate unit spherical surface images through the rotation matrix; and finally respectively mapping the twospace coordinate unit spherical surface images to obtain two panoramic coordinate images, and splicing the two panoramic coordinate images into a complete 360 DEG panoramic image. The method and system for binocular fish-eye image splicing and unfolding correction can solve the problem that the fish-eye imaging sensors are not in parallel during the assembling process, thus causing errors, and have the advantages of being good in the splicing effect after correction, being high in processing timeliness, and being easy to implement.

The invention provides a three-dimensional geometrical measurement method and a three-dimensional geometrical measurement system based on LED tag tracking. A three-dimensional scanner is arranged such that the three-dimensional scanner can move along a track and is close to an object to be measured, and blocked local measurement is flexibly and conveniently carried out on a large-scale object by a structured light method to obtain a high-resolution local image. Meanwhile, as the light of an LED tag is bright enough, even when the distance between a stereoscopic tracker and the LED tag increases, the stereoscopic tracker can quickly and accurately track the LED tag to ensure that a server can precisely locate the center position of the LED tag and the position of the LED tag in the coordinate system of the three-dimensional scanner, precisely obtain the LED tag relationship between two local measurements, accurately evaluate the attitude change of the three-dimensional scanner and realize precise stitching of local measurement results. Moreover, three-dimensional measurement of the whole contour of a large-scale object can be completed conveniently, stably and accurately without the need for tedious switching between multiple projectors.

The invention discloses a short text-oriented optimization classification method. The method comprises the following steps of: 1, obtaining an original data set and preprocessing the original data set; 2, selecting a feature item set from the preprocessed data set; 3, training the collected large-scale corpora by using a word vector tool to obtain a word vector model; 4, performing word vector representation on each feature item in the feature item set by using a word vector model, and performing primary clustering on the word vectors of the feature items to obtain a plurality of primary feature clusters; 5, performing two-stage loose clustering in each preliminary feature cluster to obtain a plurality of similar feature clusters; and 6, replacing the feature words obtained in the step 4 with the similar feature clusters obtained in the step 5, and then carrying out short text classification by using a classifier. Traditional short text classification mostly lacks semantic expression capability and is quite high in demnsion of the feature space; according to the invention, the semantic information of the short text can be expressed better, the dimension of the feature space is reduced, the precision and efficiency of short text classification are improved, and the short text classification method can be applied to short text classification tasks in various fields, such as spamshort message classification and microblog topic classification.

The invention discloses a dust removal structure of an electrostatic purifier. The dust removal structure comprises a dust collection electrode and discharge electrodes. The dust collection electrode comprises multiple dust collection electrode units. Each dust collection electrode unit is formed by inner cavities of multiple dust collection electrode single bodies and dust collection electrode enclosing cavities. Each dust collection electrode enclosing cavity is defined by the outer walls of the corresponding dust collection electrode single bodies. The dust collection electrode single bodies are connected in a buckled manner. The dust removal structure of the electrostatic purifier is formed by connecting the integrally-formed sectional material dust collection electrode single bodies in the buckled manner, so that the space utilization rate is high, more dust collection electrode units can be formed, materials are saved, and the dust removal structure is low in weight, convenient to use, easy to clean and good in electric field intensity uniformity.

A high-speed electrical data transmission system (10) includes a signal mixer (13) for receiving high-speed data from external network transceiver (12). The signal mixer (13) converts the data into an electrical current-fluctuating data signal. A signal processor (17) is located remotely of the signal mixer (13) and is connected electrically thereto by a simple / inexpensive cable (15) having only two operative conductors and receives the current-fluctuating data signal via the cable (15). The signal processor (17) converts the current-fluctuating data signal into a voltage-fluctuating data signal for distribution to a local area network.

The invention belongs to the technical field of warehouse logistics, and particularly discloses a carrying robot and a warehouse logistics system. The carrying robot comprises a movable chassis, a portal vertically arranged on the movable chassis and a goods box conveying mechanism arranged on the portal. The goods box conveying mechanism can horizontally stretch out and draw back to pick and place goods boxes on a stock container, the goods box conveying mechanism is connected with the portal through a lifting mechanism, the lifting mechanism can drive the goods box conveying mechanism to vertically ascend and descend, and the lifting mechanism is a self-driven lifting mechanism. The warehouse logistics system comprises the carrying robot. According to the carrying robot, a goods box taking method, a goods box loading method and the warehouse logistics system, the sorting efficiency and the logistics efficiency can be improved.

The invention discloses an FPGA-based photogrammetric system multi-camera dynamic synchronous exposure circuit and method. The circuit comprises a forward channel module, an FPGA processing module, a power supply module and a clock module. The forward channel module provides connection channels for all the cameras and is also connected with the FPGA processing module to transmit the acquired photographing trigger instruction of each camera. The FPGA processing module records the asynchronous time difference of the shutter action of each camera according to the acquired camera photographing trigger instruction and calculates the value of the asynchronous time difference, and delays other trigger signals according to the value of the asynchronous time difference with the last arrived shutter action signal acting as the reference so as to generate the synchronous photographing trigger signal of each camera. The power supply module supplies power to the FPGA processing module. The clock module is configured for providing a high frequency time base clock signal for the FPGA processing module.

The invention discloses a grading material specific area determining method based on three-dimensional laser scanning and the image processing technology. For a rock mass larger than 50mm, three-dimensional scanning point clouds are obtained; a rock mass three-dimensional model is established by using the scanning point clouds to obtain the volume and the surface area of the rock mass; and the specific area is the rock mass larger than 50mm is obtained by calculating based on a specific area definition. For a rock mass with the size being 5-50mm, the circumference and area are calculated by using image processing; and the specific area of the rock mass with the size being 5-50mm is calculated by using a spherical model. The specific area of a powder particle being smaller than 5mm is obtained by referring to specific area test results of existing powder materials. A rock mass size-specific area variation curve is drawn based on specific area data obtained by the three modes. A gradation distribution curve of rockburst is obtained. The rock mass size-specific area variation curve and the gradation distribution curve of the rockburst are drawn in one diagram to obtain a grading material specific area curve.

The invention provides a video stitching method and system. The video stitching method comprises an extracting and matching step, a template establishing step, an image fusion step and a processing step. The video stitching method and system disclosed by the invention have the benefits that: a video can be perfectly fused and stitched; therefore, the stitched video is clear and seamless; and thus, user requirements are satisfied.

An optical grating counterpoint splicing method belongs to the field of optics and aims at solving the problem that errors are large by an existing method that two lenticular lens optical gratings are spliced by the aid of a concave optical grating. The optical grating counterpoint splicing method comprises the steps of jointly placing n optical gratings to be spliced on a testing bar chart, using one of n optical gratings as a standard optical grating, and enabling the testing bar chart to form a standard interference fringe on the standard optical grating, respectively moving and revolving n-1 optical gratings to be spliced to enable the testing bar chart to form splicing interference fringes on the n-1 optical gratings to be spliced, and finishing counterpoint splicing of the n optical gratings when all the splicing interference fringes are identical to the standard optical grating in thickness degree and extended lines of the splicing interference fringes and an extended line of the standard optical grating are located in the same straight line. The optical grating counterpoint splicing method is used for splicing a plurality of optical gratings.