4371 results about "Image extraction" patented technology

An information input/output device having an intuitive operating feeling and improved information viewing and discriminating properties. The device comprises an superposing image extraction unit 101 extracting a portion for superpositional display from an image to output the extracted image portion as an superposing image, a mask pattern generating unit 102 generating a mask pattern, effectors 113, 114 processing the superposing image, and the mask pattern based on the effect designation information, and a base image generating unit 115 synthesizing the mask pattern image and the original image to generate a base image. The device also comprises a switcher 116, brightness/contrast controllers 117, 118 adjusting the brightness or contrast of the display image switching means 112, a control unit 111, superpositional image display unit 124 for superposed demonstration of display image planes of the displays 122, 123 and a display position adjustment mechanism 121. The display information of the image for display in superposition is demonstrated at a position which appears to be floated or recessed from the basic display plane.

According to one embodiment, an electronic apparatus comprises an image extraction module, an image list display process module, and a playback module. The image extraction module extracts a plurality of representative images from a sequence of video content data, and outputs time stamp information which is indicative of a time point at which each of the plurality of extracted representative images appears in the video content data. The image list display process module displays, on a display area, a list of representative images appearing in a period from a start position to a predetermined display constraining position of the sequence of the video content data among the plurality of extracted representative images, and constrains the display of the representative images appearing after the display constraining position among the plurality of representative images. The playback module plays back the video content data in response to input of a playback request event.

A method for combining a real space image with a virtual image, includes: causing an imaging unit to capture an image of a real space; generating an image covering a predetermined space based on a plurality of real images in the captured real space; extracting position information of a light source based on the generated image; and adding a light source or a shadow on the virtual image based on the extracted position information of the light source.

A method of combining image data from multiple frames to enhance one or more parameters of video image quality includes acquiring a first image at a first exposure duration, as well as acquiring a second image at a second exposure duration shorter than the first exposure duration and at a time just before, just after or overlapping in time with acquiring the first image, such that the first and second images include approximately a same first scene. In this way, the second image is relatively sharp and under-exposed, while the first image is relatively well-exposed and less sharp than the second image. Brightness and/or color information are extracted from the first image and applied to the second image to generate an enhanced version of the second image.

A stereoscopic image display includes a depth map extracting unit for extracting a region of interest (ROI) and a depth map from a 3D input image, an average depth for calculating unit calculating an average depth of the ROI, a depth map adjusting unit, which converts the average depth of the ROI into a screen depth value of a display panel, shifts depth values of the 3D input image in a shift direction of the average depth by a shifted amount of the average depth of the ROI, and adjusts a depth map of the 3D input image, a left/right eye image producing unit for producing left eye image data and right eye image data based on the adjusted depth map, and a display panel driver for displaying the left eye image data and the right eye image data on the display panel.

A tubular structure, such as a blood vessel, of a subject is extracted from each of a three-dimensional image representing the tubular structure and a three-dimensional intra-tubular-structure image that has been generated from plural tomographic images of the tubular structure obtained by performing tomography on the tubular structure plural times from the inside of the tubular structure along a path in the tubular structure. Further, an arbitrary range in one of the tubular structure extracted from the three-dimensional image and the tubular structure extracted from the three-dimensional intra-tubular-structure image is correlated with a corresponding range in the other one of the tubular structures. Further, a projection three-dimensional image is generated by projecting an image of a specific structure included in the range in the three-dimensional intra-tubular-structure image into the correlated range in the three-dimensional image.

A method for following hand movements in an image flow, includes receiving an image flow in real time, locating in each image in the received image flow a hand contour delimiting an image zone of the hand, extracting the postural characteristics from the image zone of the hand located in each image, and determining the hand movements in the image flow from the postural characteristics extracted from each image. The extraction of the postural characteristics of the hand in each image includes locating in the image zone of the hand the center of the palm of the hand by searching for a pixel of the image zone of the hand the furthest from the hand contour.

A computer-implemented system and method for predicting an image quality of an image are disclosed. For an input image, the method includes generating a first descriptor based on semantic content information for the image and generating a second descriptor based on aesthetic features extracted from the image. With a categorizer which has been trained to assign a quality value to an image based on first and second descriptors, a quality value is assigned to the image based on the first and second descriptors and output.

For distributing a user image corresponding to the user position in a virtual space, the invention provides a method of fetching the user image taken with a video camera into the main body of a computer (S1), judging the user position in the virtual space (S2), setting a large quantization coefficient in case of an office view (S3) but setting a small quantization coefficient and an even smaller quantization coefficients for a face area in other cases (S4), then compression encoding the image data with the designated quantization coefficient ((S5) and transferring the image data to a host server apparatus 12 through a network (S6). The user terminal receives images of other users from the host server apparatus (S7), expands the image data (S8) and displays the image data on a display apparatus 44 (S9).

An image retrieval apparatus generates an image feature list having a high and/or a low discrimination capability against other images based on the number of stored image features. Based on the image feature list, the image retrieval apparatus selects feature points extracted from a query image.

It is an object of the present invention to allow easier acquisition of image data having the necessary luminosity range. The pixel value distribution detecting unit detects the distribution of the pixel values of the input image data, constituted by pixel values that are substantially proportional to the logarithm of the quantity of incident light. The extraction range setting means sets one or more extraction ranges, which are ranges of predetermined pixel values and have the same width, on the basis of the distribution of the pixel values of the input image data. The image extraction unit produces processed image data by extracting pixels having pixel values contained in the extraction ranges from the input image data, and supplies the processed image data thus produced to the image detecting unit. The present invention can be used in image detection devices.

An apparatus and method for building a map are provided. According to the apparatus and method, a path is generated on the basis of the degrees of uncertainty of features extracted from an image obtained while a mobile robot explores unknown surroundings, and the mobile robot travels along the generated path. The path based on the degrees of uncertainty of the features is generated and this may increase the accuracy of a feature map of the mobile robot or accuracy in self localization.

The invention discloses a pedestrian re-identification method based on multi-attribute and multi-strategy fusion learning. The method of the invention includes the steps of in an offline training phase, firstly selecting pedestrian attributes which are easy to be judged and have a sufficient distinguishing degree, training a pedestrian attribute identifier on an attribute data set, then labeling attribute tags for a pedestrian re-identification data set by using the attribute identifier, and next, by combining the attributes and pedestrian identity tags, training a pedestrian re-identification model by using a strategy fused with pedestrian classification and novel constraint comparison verification; and in an online query phase, extracting features of a query image and images in a database by using the pedestrian re-identification model, and calculating the Euclidean distance between the feature of the query image and the feature of each image in the database to obtain the image with the shortest distance, which is considered as the result of pedestrian re-identification. In terms of performance, the features in the invention are distinguishable and high accuracy is obtained; and in terms of efficiency, the method of the invention can quickly search for the pedestrian indicated by the query image from the pedestrian image database.

The invention belongs to the field of depth learning, and provides a rail surface defect detection method based on the depth learning network, aiming at solving various problems existing in the priorrail detection methods. The depth learning method first automatically resets the input rail image to 416*416, and then extracts and processes the image. Image extraction mainly by Darknet-53 model complete. The processing output is mainly accomplished by the FPN-like network model. Firstly, the rail image is divided into cells. According to the position of the defects in the cells, the width, height and coordinates of the center point of the defects are calculated by dimension clustering method, and the coordinates are normalized. At the same time, we use logistic regression to predict the fraction of boundary box object, use binary cross-entropy loss to predict the category contained in the boundary box, calculate the confidence level, and then process the convolution in the output, up-sampling, network feature fusion to get the prediction results. The invention can accurately identify defects and effectively improve the detection and identification rate of rail surface defects.

An image processing method for executing edge enhancement for an original image includes: extracting edge components based upon the original image; correcting the extracted edge components by attenuating the individual edge components so that a frequency distribution related to intensity of the edge components approximates a Gaussian distribution assuming a specific width; and executing edge enhancement for the original image based upon the corrected edge components.

A method of extracting desired features from a cellular image including the steps of: (a) selecting an initial cell within the image; (b) selecting an additional cell, near the initial cell, appearing to be associated with a desired feature; (c) repeating step (b) for further cells, near at least one of the previously selected cells, appearing to be associated with said feature, until selection termination criteria are satisfied; and (d) repeating steps (a) through (c) for other initial cells. The method is particularly adept at extracting relatively weakly defined features in relatively noisy images, such as extracting faults or geologic horizons from 2D or 3D seismic data. A method of editing/filtering the features utilizing a stereo net is also disclosed. Related computer system and computer program products for implementing the method are also described.

A personal identification device of the present invention is provided with a registered data storage unit for storing registered data containing facial feature data for a registered user, a recognition parameter storage unit for storing recognition parameters, a image determination unit for determining whether or not the image quality of a user facial image input from an imaging device is appropriate, an adjustment unit for adjusting the settings of the imaging device or modifying the recognition parameters, in accordance with the result of determination carried out in the image determination unit, a feature extraction unit for extracting user facial feature data from a facial image, and a face recognition unit for comparing the extracted feature data with the registered data to determine whether or not the user is a registered user based on the result of comparison and the recognition parameters.

A method is provided for extracting a foreground object and a background sprite, wherein a provisional sprite is generated, the foreground and the background is separated on the basis of the provisional sprite, and the background sprite is generated. Another method is provided for extracting a segmentation mask by using a difference image, including a first step of regarding each of first macro-blocks as the foreground when an value of the first macro-block is larger than a first predetermined value and a second step of regarding each of second macro-blocks as the foreground when an value of the second macro-block is larger than a second predetermined value, the second macro-block being close to a macro-block which is determined as the foreground in the first step.

The present invention provides a time-series data display method and system for displaying time-series data items so that a user can grasp time more intuitively. First data associated with a desired date is retrieved and displayed, second data associated with a contiguous date is retrieved, and a display screen for the second data is displayed smaller than that for the first data in a temporal direction starting with the desired date. Third data is associated with a date contiguous to the second date retrieved, and is displayed inside the second data with a smaller display area. Also, in a hierarchical data display method and browser system, a display area is divided into an area where icons representing data items belonging to one level are displayed, and an area where child levels are displayed. As hierarchical depth increases, the data icons are made smaller and simpler. The hierarchical structure of a file system or data base can be displayed as a Venn diagram. Data items belonging to child levels are not hidden but displayed as reduced images, whereby intended data can be located effortlessly. A cutout form and image are registered mutually independently. An identifier, position, and size of the cutout form are specified as the attributes of the image. Thus, an image can be fetched into album software by performing a simple operation, or a cutout can be changed in size.

The present invention is directed to an automated vehicle identification and tracking system. The system includes at least one area monitoring system that has a plurality of imaging units disposed in an area. Each imaging unit is configured to capture an image of a monitored vehicle disposed in the area. A control system is remotely coupled to the at least one area monitoring system. The control system is configured to classify and track the monitored vehicle based on anonymous vehicle feature data extracted from the captured image. The system can assimilate eyewitness input concerning a vehicle based on anonymous vehicle feature data, including time and location of the sighting of the target vehicle, wherein the system can effectively identify the current location of candidate target vehicles within the monitored area.

Technology related to an apparatus and method for supporting a diagnosis of a lesion is provided. The apparatus includes a standard image determiner to determine, as a standard image, an image where a region of interest (ROI) is detected among a plurality of images received from an image collecting device, a candidate image extractor to extract one or more candidate images with respect to the determined standard image, a comparison image selector to select one or more comparison images for supporting a diagnosis of the ROI from the one or more candidate images, and an interface to provide a user with an output result of the standard image and the one or more selected comparison images.

The invention discloses an ATM (Automatic Teller Machine) cheat-proof device and system based on three-dimensional human face identification. The device comprises a human face detection unit and a special image processing unit, wherein the human face detection unit is used for acquiring a visible-light human face image, a three-dimensional depth human face image and an infrared human face image of a user; the special image processing unit is used for receiving the visible-light human face image, the three-dimensional depth human face image and the infrared human face image of the user, which are output by the human face detection unit, extracting corresponding characteristic information, transmitting the characteristic information to a background server for being compared with pre-stored characteristic information of the corresponding user, and receiving a comparing result returned by the background server. According to the ATM cheat-proof device and system based on the three-dimensional human face identification, ATM operators are compared and judged by using a manner of combining three types of information including the three-dimensional depth human face image, the visible-light human face image and the infrared human face image so that the operation safety of ATMs is improved.

A sound source-separating device includes a sound-collecting part, an imaging part, a sound signal-evaluating part, an image signal-evaluating part, a selection part that selects whether to estimate a sound source direction based on the first sound signal or the first image signal, a person position-estimating part that estimates a sound source direction using the first image signal, a sound source direction-estimating part that estimates a sound source direction, a sound source-separating part that extracts a second sound signal corresponding to the sound source direction from the first sound signal, an image-extracting part that extracts a second image signal of an area corresponding to the estimated sound source direction from the first image signal, and an image-combining part that changes a third image signal of an area other than the area for the second image signal and combines the third image signal with the second image signal.