376 results about "Object contour" patented technology

A user photographs an image including an object to be extracted, uses an instruction selection unit or the like to designate an extraction range, and an object extraction circuit performs extraction. The image data of an extracted object area is compressed / encoded, and stored together with photographing conditions into a storage unit. Subsequently, a background image is photographed or inputted. While the background image is displayed, the previously extracted object image is read from the storage unit. To suppress a difference in gradation and hue between the background image and the object image, the gradation and hue of the object image are adjusted, mixing and smoothing with the background image are performed in the vicinity of an object outline, and the object image is overwritten, synthesized, and displayed on the background image. The position and size of the object image are adjusted according to a user's instruction. Synthesized image data is recorded in a recording medium.

A method for providing improved foreground / background separation in a digital image of a scene is disclosed. The method comprises providing a first map comprising one or more regions provisionally defined as one of foreground or background within the digital image; and providing a subject profile corresponding to a region of interest of the digital image. The provisionally defined regions are compared with the subject profile to determine if any of the regions intersect with the profile region. The definition of one or more of the regions in the map is changed based on the comparison.

The present invention facilitates solid object reconstruction from a two-dimensional image. If an object is of known and regular shape, information about the object can be extracted from at least one view by utilizing appropriate constraints and measuring a distance between a camera and the object and / or by estimating a scale factor between a camera image and a real world image. The same device can perform both the image capture and the distance measurement or the scaling factor estimation. The following processes can be performed for object identification: parameter estimation; image enhancement; detection of line segments; aggregation of short line segments into segments; detection of proximity clusters of segments; estimation of a convex hull of at least one cluster; derivation of an object outline from the convex hull; combination of the object outline, shape constraints, and distance value.

The present invention detects a moving object by generating the distance information of the moving object, detecting the object motion, determining the object distance, detecting the object image area and the object contour from the video image that includes the object image and contour, and provides a moving object detection apparatus to carry out such detection as well as detecting a contour of the specific moving object by detecting the center of the moving object in high precision.

A method and system for biometric identification utilizes structured light (e.g., fine grid) that is projected to the target objects (fingers and palm). The image with structured light on the object contours can give 3D information of the objects with resolution to satisfy NIST fingerprint standard. In embodiments, the structured light can be of different color, to reduce the burden of the software processing. To enhance the capture speed, a model with 2D image on “smooth” 3D surface can be used. In this “2.5-D” case, the finest feature of the structured light can be as big as millimeter size. Although grid-structured light can satisfy current requirement of fingerprint and palm print capture standards, in other embodiments, a customized structure based on the initial scan result can give optimal result on the 3D data. This may be useful when higher resolution is desired.

Disclosed is a computer-implemented method of pill analysis including the steps of acquiring a pill image having an image frame and detecting contrast shifts within the image frame to locate at least one object with an object outline. A first value for the object(s) is determined, where the value is an area, a position, a length, a width, an angle, a color, a brightness, a code, a shape, a crystal pile size, a crystal geometry, a substance identity, or a character identity. Based on the first and second values, the computer outputs a result to a user.

The invention provides an industrial robot visual recognition positioning grabbing method, a computer device and a computer readable storage medium. The method comprises the steps that image contour extraction is performed on an acquired image; when the object contour information exists in the contour extraction result, positioning and identifying the target object by using an edge-based templatematching algorithm; when the target object pose information is preset target object pose information, correcting the target object pose information by using a camera calibration method; and performingcoordinate system conversion on the corrected pose information by using a hand-eye calibration method. The computer device comprises a controller, and the controller is used for implementing the industrial robot visual recognition positioning grabbing method when executing the computer program stored in the memory. A computer program is stored in the computer readable storage medium, and when thecomputer program is executed by the controller, the industrial robot visual recognition positioning grabbing method is achieved. The method provided by the invention is higher in recognition and positioning stability and precision.

Disclosed herein is a method and apparatus for editing an image using an object contour extracted from an input image. The method of editing an image using a contour-extracting algorithm includes: inputting image data; extracting an object contour from the input image data; optimizing the extracted contour using the characteristics of the input image data; editing the input image data using the optimized contour; and outputting the edited image data.

One frame of sine stripe projection in dual frequency and two-color is adopted in the method. Using Fourier transform obtains information of height of object to be measured in high and low two accuracies in measurement. Carrying out process of removing enwrapping obtains removed enwrapping value of phases of 3D body outline. Under condition of ensuring accuracy of removing enwrapping, the measuring procedure only needs to collect one frame of image so as to raise measuring speed greatly.

A system and process for tracking an object state over time using particle filter sensor fusion and a plurality of logical sensor modules is presented. This new fusion framework combines both the bottom-up and top-down approaches to sensor fusion to probabilistically fuse multiple sensing modalities. At the lower level, individual vision and audio trackers can be designed to generate effective proposals for the fuser. At the higher level, the fuser performs reliable tracking by verifying hypotheses over multiple likelihood models from multiple cues. Different from the traditional fusion algorithms, the present framework is a closed-loop system where the fuser and trackers coordinate their tracking information. Furthermore, to handle non-stationary situations, the present framework evaluates the performance of the individual trackers and dynamically updates their object states. A real-time speaker tracking system based on the proposed framework is feasible by fusing object contour, color and sound source location.

A target moving object tracking device takes a time series of picture images of a target moving object, and track a movement of the moving object in the picture images for displaying an enlarged view of the moving object. The device includes a template memory storing a template image which is compared with each one of time-series outline images given from the picture images to determine a partial area in match with the template image for each of the outline images and extract the partial areas as a moving object outline image. The template image is constantly updated by being replaced with a combination of the previous moving object outline images to accurately reflect the moving object.

The invention discloses a target recognition method based on an image contour characteristic. The method comprises the steps of preprocessing a template image and a to-be-identified object image for generating a binary image; establishing a characteristic database of an object template contour; extracting a whole contour of an object template binary image, equidistantly acquiring a certain number of characteristic points on the contour, and describing the object contour by means of a context characteristic of the characteristic points; performing target recognition on the to-be-identified object image; extracting the contour edge of the to-be-identified binary image; selecting a certain number of characteristic points; converting the shape direction of the to-be-identified object to the direction of the template; describing the contour of the to-be-identified image after direction changing by means of the context characteristic of the selected points; and determining similarity between the to-be-identified object and the template object through matching cost. Compared with prior art, the target recognition method has advantages of settling a problem of rotational invariance in a contour matching process, realizing invariable rotation in the contour matching process, and realizing effective application in target recognition in the image.

The invention provides a method of identifying large and medium-sized objects based on multi-source remote sensing image fusion. The method comprises that multispectral and SAR images are segmented into image areas according to spectrum distribution between different objects and differences of electromagnetic radiation, and a potential area of an object is extracted according to experience and knowledge of object distribution, so that characteristic extraction and object identification are more targeted while identification efficiency and identification accuracy of the system are improved; that based on the above, object areas are determined by extracting object contour, spatial layout and the like, and sub-object areas are divided according to spatial layout relation of the object contour and sub-object spatial layout; and that sub-object characteristics are extracted, and with the guidance of object characteristic set and object prior knowledge, sub-object identification and object verification are finally realizedby execution of identification rules and matching of characteristics.

The invention belongs to the field of object size measurement, in particular to an object size measurement method based on depth image processing, comprising the following steps: (1) collecting and determining a reference image, collecting a depth image of the object by using a depth camera, and calculating a gray difference value between the depth image and the reference image to obtain a difference value map; (2) constructing Otsu algorithm model, binarizing the difference map to obtain binary image; (3) using edge detection method to detect the edge of the binary image and determine the contour of the object; (4) calculating the smallest rectangle for determining the contour of the object; (5) the height of the object is obtained by using the minimum rectangle of the object contour andthe distance difference between the reference image and the camera; (6) The length and width of the object are calculated by using the principle of small hole imaging. The invention can realize the automation of the size measurement of the object, and effectively improves the speed and accuracy of the size measurement of the object.

This paper provides a new image segmentation algorithm for object-based image retrieval. The system partitions multi-dimensional images into disjoint regions of coherent color and texture. In order to distinguish the object contour lines from texture edge, the description length of the line is used as discriminating criteria. Visual attribute values are assigned to each region.

The invention relates to a binocular image and object contour-based virtual and actual sheltering treatment method, which comprises the following steps of: accurately calculating the regional contour of a real object by an interactive image segmentation algorithm, and taking the regional contour as the geometrical information of the real object; and according to the negatively-correlated relation between parallax and depth, estimating the parallax of a virtual object under a current visual angle, determining the relative depth information between the virtual object and the real object in a current scene, and hierarchically dividing the scene. The front-and-back sheltering relation between the virtual object and the real object is estimated and determined by using the relative depth information, the depth information is not required to be calculated in a pixel-by-pixel manner; and the virtual and actual sheltering treatment is achieved in a two-dimensional image space, whether the sheltering between the virtual object and the real object exists or not is judged, and corresponding treatment is carried out, so that the binocular image and object contour-based virtual and actual sheltering treatment method can be applicable to most sheltering circumstances and a better virtual and actual sheltering treatment effect is achieved. The binocular image and object contour-based virtual and actual sheltering treatment method can be widely applied to the spatial sheltering treatment of virtual reality systems, such as interactive digital entertainment, sports research and training simulation, distance education and training and the like.

A method for automatically cropping image objects, including: an object contour detecting step, an image dividing step, and an image merging step. The object contour detecting step detects object contours with Modified Laplacian Contour Detector to generate a binary image. The image dividing step divides the binary image into a background region and a plurality of non-background regions with Connected Components Algorithm (CCA). The image merging step merges the closer non-background regions from large to small, and determines whether the merging is reasonable for cropping the image object.

Method of estimation and segmentation of the volume of a three-dimensional object in medical imagery, in which:one defines a given number of base points constituting a first three-dimensional shape defined by facets;each facet of the first shape being defined by three segments, the segments are divided by defining second order points adapted to the contour of the object, so as to constitute a second three-dimensional shape closer to the contour of the object than the first shape;each segment is iteratively divided into subsegments adjusted by defining third order points adapted to the contour of the object, so as to constitute a third three-dimensional shape closer to the contour of the object than the second shape;then, the volume of the third three-dimensional shape is calculated.

A three-dimensional (3D) scanner including a light-source module, a screen, a rotary platform, an image capturing unit and a process unit is provided. The light source module is configured to emit a beam. The screen disposed on a transmission path of the beam has a projection surface facing the light source module. The rotary platform carrying a 3D object is disposed between the light source module and the screen. The 3D object is rotated to a plurality of object orientations about a rotating axis to form a plurality of object shadows on the projection surface. The image capturing unit is configured to capture the object shadows from the projection surface to obtain a plurality of object contour images. The process unit is configured to read and process the object contour images to build the digital 3D model related to the 3D object according to the object contour images.

The embodiment of the invention provides a method for tracking a video object and a device thereof, and relates to the technical field of image processing. The method and the device are invented for realizing the accurate tracking of the video object. The method comprises the steps of: picking feature points of the profile of the video object in a current frame of image; finding out matched feature points; detecting at least one alternative profile of the video object in a next frame of image; calculating the profile feature value of the video object in the current frame of image; calculating the profile feature value of the alternative profile; and comparing the profile feature value of the alternative profile with that of the video object in the current frame of image and taking the alternative profile as the profile of the video object in next frame of image if the two profile feature values are matched. The method and the device can improve the accuracy of tracking the video object.

The application provides an auxiliary shooting method and system, and relates to the technical field of an intelligent terminal. The method comprises the steps of: acquiring a reference picture; analyzing the reference picture, dividing elements in the reference picture into a scenic spot object and a figure object, and respectively drawing a scenic spot object outline and a figure object outline;displaying the scenic spot object outline and the figure object outline in the reference picture in a viewing frame; regulating a viewing direction and position, and when a overlap ratio of an actualscenic spot and the scenic spot object outline reaches a certain value, proceeding to a stage of preparing to shoot; and in the stage of preparing to shoot, when an overlap ratio of an actual figureand the figure object outline reaches a certain value, sending shooting reminding. Compared to the prior art, according to the auxiliary shooting method provided by the application, photos similar with the reference picture can be obtained, the shooting difficulty is reduced, and quality of a shot picture is improved.

A method of representing an object appearing in a still or video image, by processing signals corresponding to the image, comprises deriving the peak values in CSS space for the object outline and applying a non-linear transformation to said peak values to arrive at a representation of the outline.

The invention discloses a three-dimensional object modeling method and device, an image processing device and a medium. The method comprises the following steps: extracting a plane contour of each panoramic image in a three-dimensional space; normalizing the scale of the camera position and the scale of the plane contour of each panoramic image in the three-dimensional space to obtain the normalized plane contour of each panoramic image in the three-dimensional space; and rotating and translating the three-dimensional point coordinates of the planar contours of the panoramic images subjected to scale normalization in the three-dimensional space based on the positions of the cameras so as to unify the three-dimensional point coordinates into the same coordinate system, and splicing the object contours into a multi-object planar contour. The high-resolution panoramic images are used, so that the resolution of a model constructed for the three-dimensional object can be improved, multi-contour splicing and subsequent multi-contour optimization are carried out through coordinate transformation, the object modeling precision can be improved, and the resolution and accuracy of the generated model of the three-dimensional object can be effectively improved.

The invention discloses a target-contour three-dimensional information extraction method based on an edge curvature angle. The method comprises the following steps of (1) using a machine vision measurement technology to acquire three-dimensional point cloud data of a target object contour; (2) carrying out filtering denoising processing on the three-dimensional point cloud data; (3) ordering the denoised point cloud data through an annular ordering algorithm so that the point cloud data is arranged annularly according to an anti-clockwise direction, wherein adjacent points are stored at adjacent positions of a linear list at this time; (4) according to an annular ordering result, carrying out cluster segmentation on the point cloud data; (5) based on an edge curvature angle, rapidly dividing the point cloud data into linear points or curve points; and (6) based on point classification, carrying out segmentation fitting on the point cloud data and recovering three-dimensional information of the target contour. In the invention, the three-dimensional information of an object with a regular shape can be accurately extracted in real time.

The embodiment of the invention discloses a video processing method and server. The method includes identifying a target object from a current frame of that video and determining an object contour ofthe target object in the current frame; determining a target region in the current frame based on the object contour, and determining an image within the target region as a cropped frame of the current frame, wherein at least a portion of the object contour is contained in the target region; generating a clipped video corresponding to the video based on the clipped frame determined from the video.The technical scheme provided by the embodiment of the present application can improve the quality of the vertical video obtained after video processing.

The invention provides a people counting method based on elliptical ring template matching. The method comprises the following steps of: (S1) getting a video stream image in a monitoring area as an input image; (S2) performing edge detection on the input image through a Sobel operator to get an edge image; (S3) getting a differential image by processing through a background subtraction method according to the input image, and further performing binarization processing on the differential image to get a moving object area; (S4) combining the edge image with the moving object area, performing and operation, and extracting common parts to get moving object contours; (S5) scanning the moving object contours through a plurality of elliptical ring templates, and performing contact ratio Co calculation to extract pedestrian contours and the mass center of each pedestrian contour; and (S6) tracking and counting the mass centers based on a nearest neighbor matching and tracking method of a Kalman wave filter to get the number of people. The people counting efficiency and the accuracy are improved by matching the plurality of the elliptical ring templates with the moving object contours.