121 results about "Imaging data" patented technology

A correcting portion of an endoscope subjects a reference image to a magnification correction process so that the distance between a first mark and a second mark within the reference image is substantially equal to the distance between a first mark and a second mark within a comparative image. Next, the corrected reference image is subjected to a rotation correction process so that the orientation of the second mark with respect to the first mark within the reference image is substantially equal to the orientation of the second mark with respect to the first mark image within the comparison image. Further, the corrected reference image is subjected to a brightness distribution correction process so that the brightness distribution thereof becomes equal to that of the comparison image. The display state of the reference image and comparison image are made substantially equal, comparative diagnosis is facilitated, and diagnostic efficiency is improved.

This invention relates to optical systems for holographic projectors. We describe a holographic image projection system, the system including: a spatial light modulator (SLM) for displaying a hologram; first optics to provide an input beam to said SLM; second optics to process an output beam from said SLM to provide a displayed image; and a hologram processor to receive image data for display and to output data to said SLM to display a hologram to provide said displayed image; and wherein at least one lens of said first optics or said second optics is encoded in said hologram.

A test chart for printing black bars and white bars of a barcode, each of which has one of multiple different numbers of dots-in-width, is printed on a printing device on which barcodes are printed. In image data on the test chart, the space between neighboring black bars on the test chart is used as a white bar. The widths of black bars and white bars of a barcode are measured based on the image of the printed test chart. Based on this measurement result, the numbers of dots-in-width of black bars and white bars to be set at a printing time are calculated as barcode correction values so that the black bar widths and white bar widths of a printed barcode meet predetermined sizes. This configuration generates barcode configuration information, or barcodes, that satisfy individual user's use condition, quickly and with the minimal use of paper and ink.

The invention relates to an LED pixel correction coefficient uploading method of an LED display screen, which comprises the following steps: converting a plurality of LED pixel correction coefficients of the LED display screen into a plurality of correction image data, wherein each LED pixel correction coefficient comprises a plurality of correction information, and each correction image data comprise one or a plurality of correction information of part or all of the LED pixel correction coefficients; conveying the correction image data to a video interface, electrically connected with the LED display screen, of the control system of the LED display screen, and analyzing the LED pixel correction coefficients of the LED display screen from the correction image data through the LED display screen control system for saving. The method has the advantages that correction coefficients are converted into correction image data, and the correction image data are uploaded through the video interface, so that the uploading speed of the correction coefficients is increased.

The invention discloses a remote sensing image scene classification method based on an improved residual network, which comprises the following steps of (a) collecting and/or downloading scene imagesto obtain a remote sensing scene classification data set; (b) performing enhancement processing on the images in the remote sensing scene classification data set, and randomly dividing the images in the remote sensing scene classification data set into a training set and a test set; (c) improving the residual network structure, training the improved residual network by using the images in the training set, continuously optimizing the residual network structure by training, and testing the accuracy of the residual network by using the images in the test set after the training is completed; and(d) carrying out scene classification on the remote sensing images by using the trained residual network structure. According to the method, the scene classification is carried out on the remote sensing image by utilizing the improved residual network, the network is more suitable for obtaining the remote sensing image data set with large difficulty and complex background textures by adjusting thestructure of the residual network, and the classification accuracy is improved.

The invention discloses a tumor MRI weak supervised learning analysis modeling method, which comprises the following steps: S1, constructing a tumor MRI segmentation network based on a full convolutional neural network to realize tumor MRI preliminary segmentation; S2, taking the tumor MRI segmentation network as a generator, taking the classification model as a discriminator, and proposing a generative adversarial training algorithm of the tumor MRI segmentation network; and S3, taking unlabeled image data in the medical image as additional input condition information of the generator model and the discriminator model, guiding a data generation process, and enhancing the stability of the segmentation-generative adversarial network. The invention also discloses a model established by usingthe tumor MRI weak supervised learning analysis modeling method. According to the method, the tumor target region and the endangered organ can be more accurately and automatically segmented under thecondition of fewer training samples, and the method is applied to target region and endangered organ sketching of tumor radiotherapy.

The invention relates to a deep neural network target detection method integrated with a pixel-level attention mechanism. The deep neural network target detection method comprises the following steps:collecting a training sample image; converting data in the collected image data set into a format required by training the deep convolutional neural network through preprocessing, and performing dataenhancement to improve the training effect of the neural network; designing a deep convolutional neural network structure integrated with a pixel-level attention mechanism, the network structure being used for receiving an input image and outputting bounding box regression information and category information to an object in the image: the network structure comprising two parts, one part being abasic network for preliminarily extracting features; the other part being a plurality of cascaded convolutional layers or downsampling layers added at the tail end of the basic network and being usedfor extracting convolutional feature maps with different scales and fusing the multi-scale feature maps with a pixel-level attention mechanism; and training.

The invention provides a gender classification method and a system thereof based on facial images. The gender classification method includes obtaining and pre-processing facial images to be classified to generate image data to be classified, obtaining local circle pattern (LCP) code of each pixel point of image data to be classified except edge pixel points to gain LCP code to be classified, obtaining LCP code of each pixel point according to gray value difference of other pixel points in area near the pixel point area, and determining classifications of image data to be classified according to LCP code images to be classified and pre-trained classification modules. In the process of determining LCP code, and different values of adjacent pixel points and a center pixel point are considered, that is, difference in symbols and difference in size are considered, therefore, feature description of image data is enhanced, and accuracy of gender classification by computer according to facial images is improved.

A moving image reproducing apparatus which receives moving image data transmitted through a network and reproduces the moving image data includes an analysis unit adapted to analyze first moving image data stored in a storage unit and additional information of the first moving image data, a determination unit adapted to determine a storage destination on the network which is configured to acquire second moving image data following the first moving image data on the basis of the analysis result on the additional information, an acquisition unit adapted to acquire the second moving image data from the storage destination of the second moving image data, and a reproducing unit adapted to reproduce the second moving image data acquired by the acquisition unit concurrently with finishing reproduction of the first moving image data.

The invention discloses an image processing method and device. The image processing method comprises the following steps: reading a to-be-synthesized image; reading a to-be-cutout image, performing RGB channel splitting on the to-be-cutout image and respectively calculating edge pixel average values of the three channel images; performing binarization processing on the images of the three channelsby using the average values, and combining binarization images of the three channels to obtain a superposed binarized image so as to record target profile information of the binarized image; readingthe target profile information to generate a position of a synthetic image in the to-be-synthesized image; putting the to-be-cutout image on the position of the synthetic image so as to generate the synthetic image, and saving target categories and the generated positions as tag files, wherein numbers of the target object categories and names of the target object categories in the to-be-cutout image are configured in advance, and the numbers of the categories correspond to the names of the categories. According to the image processing method and device disclosed by the invention, by utilizingimage cutout and synthesis, image data required to be learned deeply are acquired and automatic generation of the corresponding labels is realized.

One or more embodiments of the invention provide a system and method for processing an infrared image to receive infrared image data of a non-uniform scene, apply column noise filter correction to the infrared image data, and/or apply row noise filter correction to the infrared image data, and provide corrected infrared image data based on the column and/or row noise filter correction as infrared output image data.

The invention discloses a photographic processing method, a photographic processing system and a mobile terminal. The method comprises the following steps of recording a photo area, which is to be stored after photographing, selected from a photographic previewing screen by a user in advance, calculating the position and the size of the photo area which is to be stored after photographing in the whole previewed photo, and storing position and size data; during photographing, processing the image data of the selected photo area in a shot photo to generate a corresponding photo according to the position and the size of the photo area in the whole previewed photo. According to the method, the system and the mobile terminal, a new function of arbitrarily selecting the size of the photo during photographing is added to the mobile terminal; the user can arbitrarily set the size of the shot photo within a hardware-permitted range, so that convenience is provided for the user.

Complex magnetic resonance (MR) image data, acquired from an examination subject that has at least two different spin species, are processed in a method and a device. At least one MR parameter that influences data acquired in an MR data acquisition is determined with spatial resolution. Complex image data are respectively acquired with each of multiple acquisition units in an MR data acquisition at multiple echo times, such that a phase difference between the first spin species and the second spin species is different at least two of the echo times. For multiple image points, a value of the at least one MR parameter is determined so that a function that depends on the at least one MR parameter and the corresponding image point in the complex image data satisfies an extreme condition.

The application relates to the technical field of computer vision, and discloses an obstacle avoidance reminding method, a related device and a computer readable storage medium. According to the obstacle avoidance reminding method, the related device and the computer readable storage medium, ground detection is carried out according to acquired image data to acquire ground information of a road; according to the acquired ground information, the traffic condition of the road is determined; if the road is determined to be impassable, road condition detection is carried out on the road to obtaina first detection result, and obstacle detection is carried out on the road to obtain a second detection result; and according to the first detection result and the second detection result, obstacle avoidance reminding information is determined. Ground detection is carried out through the acquired image data, path detection is carried out on the basis of the ground detection, and road condition detection and obstacle detection are carried out under the condition that the road is determined to be impassable to determine the obstacle avoidance reminding information so that users can quickly andsafely pass through the obstacle.

A liquid crystal display device is provided which is capable of reducing EMI (ElectraMagnetic Interference) noises while simultaneously responding to requirements for the high-speed transmission of image data, miniaturization and thinning of a signal processing board. A timing controller outputs, in accordance with an input data signal and input clock signal, a data line driving circuit controlling signal, internal data signal, internal clock signal to a data line driving circuit and outputs a scanning line driving circuit controlling signal to a scanning line driving circuit. The timing controller has a clock signal frequency setting mode in which a frequency of each of clock signals is set to a different value and the clock signals are supplied to the data line driving circuits and other data line driving circuits in one region and another region.