30results about How to "Improve alignment quality" patented technology

An apparatus and method for forming liquid spinning solution into a solid formed product whereby the solution is passed through at least one tubular passage (17) having walls formed at least partly of semipermeable and/or porous material. The semipermeable and/or porous material allows parameters, such as the concentration of hydrogen ions, water, salts and low molecular weight, of the liquid spinning solution to be altered as the spinning solution passes through the tubular passage(s).

A liquid crystal display (200) includes a first base plate (202), a second base plate (220), a liquid crystal layer (238) located between the first base plate and the second base plate, and two alignment films (210, 228). The two alignment films are respectively positioned on inner surfaces of the first base plate and the second base plate. Each alignment film is a film of oriented carbon nanotubes (212, 230). An oriented direction of the carbon nanotubes on the first base plate is perpendicular to that of the carbon nanotubes on the second base plate. A manufacturing method for the liquid crystal display is also disclosed.

Grapheme-to-phoneme alignment quality is improved by introducing a first preliminary alignment step, followed by an enlargement step of the grapheme-set and phoneme-set, and a second alignment step based on the previously enlarged grapheme/phoneme sets. During the enlargement step, grapheme clusters and phoneme clusters are generated that become members of a new grapheme and phoneme set. The new elements are chosen using statistical information calculated using the results of the first alignment step. The enlarged sets are the new grapheme and phoneme alphabet used for the second alignment step. The lexicon is rewritten using this new alphabet before starting with the second alignment step that produces the final result.

The embodiment of the invention provides an image processing method and electronic equipment, which are applied to the technical field of communication and are used for solving the problem of poor image effect when images containing portraits are processed in the prior art. The method comprises the following steps: performing portrait segmentation on a target image frame meeting a preset conditionin N first image frames acquired by a camera for the same shooting object to obtain a portrait mask of the shooting object; performing image segmentation on the N first image frames according to theportrait mask to obtain N first portrait image frames and N first background image frames, and combining one first portrait image frame and one first background image frame to form a first image frame; aligning the N first portrait image frames and the N first background image frames; and carrying out image fusion on each aligned first portrait image frame and the corresponding aligned first background image frame, and outputting a multi-frame registered target image. The embodiment of the invention is applied to electronic equipment to shoot portrait scenes.

The invention relates to a cloth winding barrel for spinning. The cloth winding barrel comprises a rotating shaft and a hollow winding barrel body capable of rotating along with the rotating shaft. The cloth winding barrel is characterized by further comprising a cloth clamp contained in the winding drum body, the cloth clamp comprises a central rod parallel to the axis of the winding drum body, and two clamping pieces sleeving on the central rod and capable of opening and closing around the central rod, the winding drum body axially comprises an open end and a closed end, a narrow notch allowing cloth to penetrate is arranged on the side wall of the winding drum body, one end of the notch is open at the open end, a clamping groove with an inward opening is arranged on the closed end of the winding drum body, and the corresponding end of the central rod is clamped in the clamping groove. The winding drum body is arranged to be of a hollow structure and matched with the special cloth clamp for use, so that firmness in fixing of the cloth is greatly improved; the cloth winding barrel is convenient to use and small in difficulty in operating; due to high stability in fixing of the cloth, probability of deviation of the cloth in the process of winding is greatly lowered, and winding quality is guaranteed.

The invention discloses a cross-language knowledge graph-oriented Chinese-Thai entity alignment method, which is characterized by comprising the following steps of: (1) bilingual data set acquisition;(2) machine translation model constructing and training; (3) entity extraction; and (4) entity translation and matching. The method can achieve bilingual entity alignment more effectively and more accurately, and solves the problem that the entity alignment degree of cross-language knowledge graph construction is low at present.

The present invention provides a method for genotyping alleles in at least one homologous genetic loci set, comprising: (i) providing a DNA-containing sample that includes said at least one homologous genetic loci set; (ii) performing PCR amplification of regions of said homologous genetic loci set using consensus sequence-specific primers, wherein said consensus sequence-specific primers bind to consensus sequences that are common to a plurality of genes within the genetic loci set, thereby generating a pool of amplification products; (iii) sequencing a plurality of said amplification products in order to determine the relative proportion of each nucleotide at each position in a sequencing read; (iv) performing a sequence alignment between the sequencing read results of (iii) and at least one reference sequence, which reference sequence corresponds to one of the genes in said homologous genetic loci set; and (v) performing genotype calling of the allele or alleles in said sample based on the relative proportion of each nucleotide at each of a plurality of discriminant positions in said alignment. Also disclose are related products, kits and systems for performing the method.

The invention relates to an optical flow-based gigapixel video alignment method and device and a medium, and relates to a billion pixel video technology. The gigapixel video alignment method based onoptical flow comprises the following steps: determining a global image block corresponding to a local image in to-be-processed video data; inputting the local image and the corresponding global imageblock into a deep learning network to obtain a super-resolution region image, the deep learning network being used for performing optical flow estimation, optical flow super-resolution and image deformation processing on the input image; superposing the obtained super-resolution region image on the corresponding image region of the corresponding global image to obtain a super-resolution global image; and generating a gigapixel video according to the super-resolution global image. According to the invention, the deformed optical flow is guided pixel by pixel and a deep learning network based onbig data is utilized to realize fine difference resolution and alignment between images of different viewpoints, and the alignment quality of a billion-pixel video is improved.

The invention relates to pressure heads and a straightening device. The straightening device comprises an upper die and a lower die; the lower die is provided with an elongated slot for placing materials; the upper die is provided with the multiple pressure heads sequentially arranged in the length direction of the elongated slot; each pressure head comprises a sleeve base and a pair of slide blocks; all the slide blocks are slidably installed on the sleeve bases; an opening is formed between every two slide blocks; each sleeve base is provided with a driving device for pushing the corresponding slide blocks to slide up and down in the axial direction of the sleeve base; and when the slide blocks slide downwards in the axial direction of the sleeve bases, the openings are narrowed down gradually. According to the pressure heads and the straightening device, the structure is simple, and straightening can be conducted on a rod piece, especially, a copper piece; and the straightening quality is high, and the outer surface of the rod piece cannot be damaged.

A liquid crystal display is provided, which includes first and second substrates, a liquid crystal layer formed between the first and second substrates, an alignment layer formed on at least one of the first and second substrates, and a blocking layer formed between at least one of the first and second substrates and the alignment layer and including a water soluble conductive polymer and a clay.

The invention discloses a straightening machine roller carrier with a bending detection function. The straightening machine roller carrier comprises two fixing supports, positioning faces are arranged on the fixing supports, lifting frames for lifting pull rods are arranged on the side faces of the fixing supports, one lifting frame is a driving end, and the other lifting frame is a driven end; rotating wheels for driving the pull rods to rotate are mounted on the lifting frames; rotating wheels and worm gears on the driving end lifting frame is coaxially connected with each other, the worm gears are driven by worms, and the worms are driven by a first power mechanism; a bending detection mechanism for detecting the bending of the pulls rod is arranged between the two fixing supports; and the bending detection mechanism comprises a support frame, a sleeve is mounted on the support frame, a piston and an electronic ruler are mounted on the sleeve, the electronic ruler is fixed with the rear end of the sleeve, the front end of the electronic ruler extends into the rear end of the piston, a bump is arranged in the middle of the piston, a spring is sleeved on the piston between the bump and the rear end of the sleeve, and the front end of the piston is matched with the outer walls of the pull rods. The straightening machine roller carrier can realize automatic detection of rotation, lifting and bending of the pull rods, improves the straightening quality and the working efficiency of the pull rods, and reduces the operation intensity.

The invention discloses production auxiliary equipment for aluminum profile machining. The production auxiliary equipment comprises a workbench, wherein the top of the workbench is provided with positioning pieces, a preheating cylinder, a heating box, a straightening mechanism and a cooling box, and the two positioning pieces are symmetrically arranged on the two sides of the top of the workbench and serve as the positioning piece at the feeding end and the positioning piece at the discharging end; and the straightening mechanism is arranged in the middle portion of the top face of the workbench and located between the heating box and the cooling box, the preheating box is close to the feeding end of the workbench, and the cooling box is close to the discharging end of the workbench. According to the production auxiliary equipment for aluminum profile machining, aluminum profiles are straightened through a plurality of straightening discs in the straightening mechanism, the straightening discs are composed of fixed discs and mold discs, the mold discs and the fixed discs are detachably connected and are convenient to replace, the corresponding mold discs can be replaced when the profiles with different sizes and shapes are straightened, the profiles enter mold grooves to be straightened during operation, and straightening convex strips in the mold grooves can repair the groove walls of grooves in the surfaces of the profiles, so that the straightening quality is improved.

The invention provides a corpus alignment method and device, electronic equipment, a computer readable storage medium and a computer program product, and relates to the technical field of artificial intelligence such as machine translation, natural language processing and deep learning. One specific embodiment of the method comprises the following steps: after obtaining an initial corpus pair composed of a first corpus and a second corpus which are different in language, determining the initial corpus pair of which the forward conditional probability and the reverse conditional probability are greater than a first threshold value as a target corpus pair; and determining the semantic matching probability of the target corpus pair by using a preset translation model, and finally, performing corpus alignment processing on the first corpus and the second corpus in the target corpus pair of which the semantic matching probability is greater than a second threshold. According to the embodiment, the probability threshold requirement when the matching pairs are screened based on the statistical probability can be reduced, so that more word pairs which are not significant on the statistical level are recalled on the premise of ensuring the accuracy of the semantic correspondence between the first corpus and the second corpus, and the alignment quality of the corpus is improved.

The invention provides an alignment mark for contact hole alignment and a forming method thereof, and relates to the technical field of semiconductors. The forming method of the alignment mark for contact hole alignment comprises a step of forming shallow trench isolations on the two sides of an area, in which an alignment mark for contact hole alignment is to be formed, on a semiconductor substrate, and a step of forming an alignment mark of the same material as the semiconductor substrate in the area, in which an alignment mark for contact hole alignment is to be formed, on the semiconductor substrate. Therefore, the effective thickness of the alignment mark can be improved, and the quality of alignment can be improved. The alignment mark for contact hole alignment is made of the same material as the semiconductor substrate, and the shallow trench isolations are formed in the areas on the two sides of the alignment mark in the semiconductor substrate, so that the effective thickness of the alignment mark is improved, and the quality of alignment is improved.