88results about How to "Same resolution" patented technology

The invention provides an unsupervised domain-adaptive brain tumor semantic segmentation method based on deep adversarial learning. The method comprises the steps of deep coding-decoding full-convolution network segmentation system model setup, domain discriminator network model setup, segmentation system pre-training and parameter optimization, adversarial training and target domain feature extractor parameter optimization and target domain MRI brain tumor automatic semantic segmentation. According to the method, high-level semantic features and low-level detailed features are utilized to jointly predict pixel tags by the adoption of a deep coding-decoding full-convolution network modeling segmentation system, a domain discriminator network is adopted to guide a segmentation model to learn domain-invariable features and a strong generalization segmentation function through adversarial learning, a data distribution difference between a source domain and a target domain is minimized indirectly, and a learned segmentation system has the same segmentation precision in the target domain as in the source domain. Therefore, the cross-domain generalization performance of the MRI brain tumor full-automatic semantic segmentation method is improved, and unsupervised cross-domain adaptive MRI brain tumor precise segmentation is realized.

A process for forming a three-dimensional article by stereolithography, said process comprising the steps: 1) coating a thin layer of a liquid radiation-curable composition onto a surface said composition including at least one filler comprising silica-type nano-particles suspended in the radiation-curable composition: 2) exposing said thin layer imagewise to actinic radiation to form an imaged cross-section, wherein the radiation is of sufficient intensity to cause substantial curing of the thin layer in the exposed areas; 3) coating a thin layer of the composition onto the previously exposed imaged cross-section; 4) exposing said thin layer from step (3) imagewise to actinic radiation to form an additional imaged cross-section, wherein the radiation is of sufficient intensity to cause substantial curing of the thin layer in the exposed areas and to cause adhesion to the previously exposed imaged cross-section; 5) repeating steps (3) and (4) a sufficient number of times in order to build up the three-dimensional article.

An image processing apparatus and a recording medium which can improve the quality of a color document image are provided. An input image is converted into an image having low resolution. A subtractive color image is generated using the converted image having low resolution. From the subtractive color image, adjacent pixels which are allocated to the same representative color are unified so as to extract an FCR (Flat Color Region) candidate region. A region candidate is redetermined using an image having resolution which is higher than the subtractive color image. Next, a boundary of the FCR is detected and an ultimate FCR is determined. Selection of a representative color of the determined FCR is carried out, and a specific color processing which replaces a color which is close to a pure color with the pure color is effected. Finally, image regeneration is carried out by overwriting and drawing (synthesizing) the FCR on the input image.

An angle absolute encoder comprises an encoded code rod that has code marks, such as pairs of fine code lines and coarse code lines. Light is reflected from or transmitted through the code marks and detected by a light detector to determine absolute angle position.

The invention discloses a microphone array and a method for realizing the speech signal enhancement and is characterized in that firstly N matrix elements with factors in the same unit direction are uniformly and radially arranged outward on a circle with the radius of R; a circular array with radian ranges of 2Pi for receiving and treating signals is formed; or the matrix elements are uniformly and radially arranged outward on a passage of circular arc with the radius of R and a central angle of Phi to form part of a circular array for receiving and treating signals with the radian ranges of Phi; the weighted value of each matrix element and weighted vector corresponding with the incident direction in the radian ranges of 2Pi/N are calculated according to the center frequencies of audio signals of the array to be treated based on the expected signal doa; then weighted vectors of the whole circular array or part of the circular array or other incident directions is obtained by making use of cyclic shifting; finally, signals received on each matrix element are weighted and combined according to calculated weighted values.

A touch panel comprises a substrate, a touch sensing layer and a transparent cover plate. The touch sensing layer is disposed on the substrate. The transparent cover plate is of different thicknesses and is disposed on the touch sensing layer. The touch sensing layer comprises a plurality of touch units; each touch unit corresponds to single thickness of the transparent cover plate; according to the single thickness of the transparent cover plate, area of each touch unit and spacing between the touch unit and the adjacent one are set.

The invention provides a human and scene image synthesis method based on rotary camera lens photographing. The method includes the steps that a rotary camera lens is set to be in a front preview state, a self-photographing image is obtained in the preview state and a figure image is extracted; the rotary camera lens is set to be in a rear photographing state and photographs a scene image; the figure image and the scene image are synthesized to obtain a human and scene synthesis image. The invention further provides a human and scene image synthesis system based on rotary camera lens photographing. Due to the fact that the synthesized images are photographed through the same camera lens, image resolutions are the same, so that synthesis efficiency can be improved, the phenomenon that the human and scene synthesis image is distorted after figures are processed is avoided, and the synthesis effect is improved.

Provided is a manufacturing method of cold-rolled SUS304 two-sided bright-annealed (BA) surface stainless steel. The manufacturing method comprises the steps that a black rust hot roll is subjected toannealing and acid pickling; a multi-roller rolling machine carries out rough rolling, specifically, the reduction rate of rough rolling is controlled to be 10%-30%; a plurality of grinding racks areadopted to grind the upper surface and the lower surface of a rough-rolled steel plate; the multi-roller rolling machine carries out finish rolling, specifically, the steel plate is rolled to have the needed thickness; bright annealing is carried out under the whole hydrogen protective atmosphere, specifically, the annealing temperature is 1020-1140 DEG C; and after annealing under the whole hydrogen protective atmosphere is completed, modulation treatment is carried out through a leveling roller, the plate shape is adjusted, and surface glossiness of the product is improved. According to themanufacturing method, the process is reasonable, fine surface quality of the SUS304 BA product is ensured through range control of shot blasting process parameters, the ratio of abrasive belts of thegrinding process and surface control over the rollers of the rolling machine; it is ensured that the obverse side and the reverse side of the steel plate have the coincident resolution by carrying out rough grinding and finish rolling on the both sides; scratch defects of the reverse side are controlled through selection or protection of a plate shape roll of the rolling machine; and as for the prepared two-sided BA surface stainless steel, the obverse side and the reverse side have less defects and the coincident glossiness, and the two-sided BA surface stainless steel has broad applicationprospects.

The embodiment of the invention discloses two-dimension code generation method, a two-dimension use method, a two-dimension code generator and user equipment. Relatively small two-dimension codes with same contents and resolution ratios can be obtained. The use method disclosed by the embodiment of the invention comprises the following steps: obtaining a target URL (uniform resource locator); converting the target URL into a short link URL; converting the short link URL into the two-dimension codes. According to the method disclosed by the embodiment of the invention, the relatively small two-dimension codes with the same contents and resolution ratios can be generated.

The present invention discloses a display panel. The display panel comprises a bending region and a non-bending region connected with the bending region, and the display panel further comprises an array substrate. The array substrate comprises: a plurality of first pixel regions which are mutually arranged in parallel in multiple lines and arranged in the bending region; a plurality of second pixel regions which are mutually arranged in parallel in multiple lines and arranged in the non-bending region; and a plurality of flexible connection regions configured to connect with adjacent first pixel regions and connect with the first pixel regions and the second pixel regions, wherein each flexible connection region is internally provided with a slot, and each slot is filled with an organic photoresist layer. According to the invention, the slots are arranged in the flexible connection regions of the bending region to form an inorganic layer slot between two adjacent lines of pixel regionsin the bending region to separate from each two lines of pixel regions, and the slots are filled with the organic photoresist layers so as to effectively relieve the stress of each inorganic layer ormetal wiring corresponding in the display panel, especially in the array substrate.

Provided is a method for obtaining remote-sensing images by a same-sortie variable-flight-height unmanned aerial vehicle. The method comprises an unmanned aerial vehicle body, an unmanned aerial vehicle flight height control module and aerial photography equipment carried by the unmanned aerial vehicle body. The method is characterized in that in the flying process of the unmanned aerial vehicle, the flight height of the unmanned aerial vehicle is dynamically adjusted according to changes of heights of terrains. The method has the advantages that under the condition that the take-off and landing times are reduced, the situation that the ground resolution of the images is in a good state is effectively guaranteed, meanwhile, consistency of the ground resolution of the different images is good, and processing difficulty of the images in the later period is reduced.

The invention provides a target classification method based on multi-modal data characteristics. The method comprises the following steps: collecting laser radar point cloud data and RGB image data; acquiring a plurality of features of the laser radar point cloud data; obtaining a plurality of laser radar feature maps according to the plurality of features; performing up-sampling and densificationon the laser radar feature map; acquiring three-channel data of the laser radar feature map, and fusing the three-channel data with the RGB image data to form six-channel data; and training the six-channel data by adopting a deep learning network model to obtain a classification result. According to the invention, three features obtained from a laser radar are used as three-channel data and are fused with three-channel data of an RGB image to form six-channel data; six-channel data are trained by using a deep learning network model, a plurality of different probabilities can be obtained for each target after training, and a numerical value with the maximum probability is selected from the plurality of probabilities to serve as a final target classification result, so that the target can be detected more accurately, and the accuracy is higher.

The embodiment of the invention provides a display panel and a display device. A display area of the display panel comprises a first display area and a second display area. The display panel comprisesa substrate, a display layer, a state switching layer, an optical element and a light homogenizing structure layer. The display layer comprises a pixel definition layer and a plurality of light-emitting devices. A first light-emitting device located in the first display area comprises a first anode, a second light-emitting device located in the second display area comprises a second anode, and the light transmittance of the first anode is larger than that of the second anode. The state switching layer is located on the side, away from the display surface of the display panel, of the display layer, the state switching layer is overlapped with the first light-emitting device, and the light transmittance of the state switching layer in a first state is smaller than that of the state switching layer in a second state. The optical element is located on the side, away from the display layer, of the state switching layer. The light homogenizing structure layer is located between the state switching layer and the optical element and used for diverging light emitted by the state switching layer. The quantity of light received by the optical element can be improved, the problem that the light quantity is uneven after light penetrates through the first display area is solved, and the imaging quality is improved.

The invention discloses a large view field push-broom constant ground element remote sensing camera system. The system comprises the following parts: a large view field camera is formed by splicing five sub-cameras, each sub-camera has the same view field and different instantaneous view field, and the instantaneous view field of each sub-camera changes with the field angle, wherein the sub-camera C images a sub-satellite point, the sub-camera A and E, B and D are respectively symmetrically arranged on two sides of the sub-camera C, the sub-camera A and E are formed by a zoom optical system and a CCD detector with a constant pixel size, and the sub-camera B, C and D are formed by a fixed focus optical system and a CCD detector with pixel in gradually varied size. The advantages of the camera system comprises: the camera has large view field, the ground coverage area is wide, and the ground resolution is constant and consistent in whole view field range.

A single-panel color image display apparatus realizing a wide color gamut includes a light source, a colored light separator, and a light valve. The colored light separator has four or more dichroic filters of reflective type which separate a light emitted from the light source according to a wavelength. The light valve controls the light that is emitted from the light source and separated by the colored light separator according to color, on a pixel-by-pixel basis according to an input image signal, and forms a color image. By performing color scrolling using a scrolling unit having a spiral array of lens cells, the same resolution as that obtained when using a color wheel can be realized and the same light efficiency as that obtained when using a three-panel color image display apparatus can be realized.

The invention relates to a display system which image in the sight of bare eye is tridimensional. This invention includes a display. It is characterized in that it has a prominentia raster slice on the surface of the invention. This invention uses a set of oblique arranged prominentia lens array, through the refraction in horizontal direction, offers different perspective image to each eye. In this way, it can present clear tridimensional image, wipe off 70% Moore veins and angle of sight can be enlarged to 120degree,besides, the resolution factor in the direction of X and Y is quite equal. Excellence of this invention are that it can make people looking at three-dimension images without glasses and definition of images can be at least 1000 as well as extensive angle of view, many clear spots, stable brightness, and so on.