1413 results about "Computer image" patented technology

A system and method use computer image processing for setting the parameters for an image acquisition device of a camera. The image parameters are set automatically by analyzing the image to be taken, and setting the controls according to the subject matter, in the same manner as an expert photographer would be able to. The system can run in a fully automatic mode choosing the best image parameters, or a "guided mode" where the user is prompted with choices where a number of alternate settings would be reasonable.

The invention relates to computer graphics and computer imaging on a video display, and includes the dynamic detection of video windows and graphical images overlapping one another. A display processor identifies differences between typical video and graphics data sources to detect the edges of video windows. By detecting the edges of active video windows within a graphics image, a display processor may uniquely adjust image characteristics of an exposed video window. These characteristics include, for example, hue, brightness, intensity and contrast.

A system and method for using computer image processing for selectable task-based digital video compression is described. The system is intended to reduce travel of experts and let these experts direct field agents in performing tasks remotely. The tasks to be performed in the remote field can be monitoring, manipulating, and navigating. A field agent performs the manipulation and navigation operations, this agent can be a human being or a robot. The task-based compression algorithms use computer vision techniques to extract the bare minimum amount of information from the remote field scenery to allow the task to be performed. High frame rate photo-realistic reconstruction of the remote scene is not generally necessary.

The invention provides a method for barrier perception based on airborne binocular vision. The method comprises the following steps of setting a coordinate system of an airborne binocular vision camera, calculating a formula for conversion between the coordinate system and computer image coordinates of an image obtained by the airborne binocular vision camera according to the coordinate system, wherein the airborne binocular vision camera comprises a left camera and a right camera, extracting characteristic points of an image obtained by the airborne binocular vision camera, carrying out a characteristic vector description process on the characteristic points, carrying out stereo matching of left and right images according to characteristic vectors of the characteristic points to obtain preliminary matching point pairs, eliminating error matching in the preliminary matching point pairs to obtain final matching point pairs, creating a disparity map according to the final matching point pairs, and carrying out barrier perception according to the disparity map. The method for barrier perception based on airborne binocular vision has the advantages of strong adaptability, good instantaneity and good concealment performance.

A system and method related thereto, for automatically inspecting at least one tire of a moving vehicle. The system includes an image acquisition device including at least one camera. The camera acquires multiple real-time images of at least one tire of a moving vehicle. As the tire of the vehicle enters a field of view of the camera, the camera component is configured to photograph and transmit the multiple photographs of the at least one tire. A light source illuminates the at least one tire of the vehicle in the field of view of the camera, and the light application may be synchronous with the imaging performed by the camera. A computer image processing component receives the multiple images transmitted from the camera and analyzes the multiple images using a detection program. The computer imaging processing component analyzing the photographs and provides an inspection analysis of each at least one tire.

The invention provides an automatic recognition and an extraction method and a device for a bill information area and relates to the fields of computer image processing and information recognition. According to the automatic recognition and the extraction method for the information area of the bill, processes of the information recognition, edge and shading processing, character zone enhancement processing, information area extraction and the like are performed through bill image preprocessing and according to an information area character database which is built in advance through self-learning and training of the information area. The automatic recognition and the extraction method for the bill information area has the advantage of being capable of automatically recognizing and extracting interested area information of multiple bills with good accuracy and adaptability to save manpower and time of the bill processing.

A single integrated circuit includes first and second data processors operating on different instruction sets independently operating on disjoint programs and data. The single integrated circuit preferably includes an external interface, a shared data transfer controller and shared memory divided into plural independently accessible memory banks. The two data processors are preferably a digital signal processor (DSP) and a reduced instruction set computer (RISC) processor. The DSP and RISC processors are suitably programmed to perform differing aspects of computer image processing.

The invention relates to a traffic signal light intelligent control system and a control method thereof, which belong to the technical field of intelligent traffic; the system comprises four CCD camera heads, a signal processor and a controller; and the method comprises the following steps: establishing a coordinate system and the transformation relation thereof; sequentially carrying out distortion correction, sharpening and perspective transformation on a traffic flow image; extracting edges of a vehicle image and filtering out lane lines and traffic sign arrows in the image; cutting the image; carrying out edge projection for obtaining pixel coordinates which correspond to the queue length of vehicles waiting to pass on each type of lane in each direction under the computer image coordinate system; calculating the actual queue length of the vehicles waiting to pass; obtaining time required for leading all the vehicles on each type of the lane in each direction to pass; and controlling the opening and the closing of a traffic signal light according to the time of leading all the vehicles to pass. The traffic signal light intelligent control system can control the opening time and the closing time of the traffic signal light in a high-efficient and energy-saving manner, thereby providing a scientific information base for realizing the urban intelligent traffic.

The invention discloses a handwriting generating method for an electronic writing brush, which relates to the technical field of computer image processing and includes the following steps: S1 calculating distance between two adjacent touch points according to touch point coordinates on a touch track delivered by a touch device; S2 if the distance is smaller than a preset threshold, the touch point is obtained according to the preset threshold serving as the distance, and if the distance is larger than the preset threshold, an interpolation point is inserted between the two adjacent touch points; and S3 drawing a preset writing brush shape according to points on the touch track to generate handwriting. The handwriting generating method enables writing brush handwriting input by hands to be smooth and continuous.

The method includes following steps: cyclic variational dynamic markers recognizable to computer image system are installed on moving targets; based on realtime graphic shot by panorama pick-up head, image analysis module finds out position data in coordinates of dynamic markers in whole image, and sends the data to the capture module for moving target; the capture module calculates specific positions of the dynamic markers within range of panorama pick-up head, as well as controls rotation of tracking pick-up head; thus, shooting focus aims at specific positions of the dynamic markers in realtime; sending images of the dynamic markers taken in realtime to pointed monitor for display. Using current equipment, the invention increases intelligence and controllability. Moreover, ordinary user can install system by self to monitor dynamic object expediently without need of help from specialized technical personnel.

This invention relates to one visual frequency stable method based on property matching in the computer image visual frequency process technique field, which comprises the following steps: finding out each frame of visual SIFT property points; adopting simulation module as parameters estimation module for total estimation; adopting Gauss filter and curve matching method to process visual sequence process to fill unknown area.

The invention provides an intelligentized precision automatic hay mower and a control device thereof. According to the intelligentized precision automatic hay mower and the control device thereof, a direct current accumulator serves as the power source, computer image processing, distributed beacon location, laser scanning displacement and angle measurement, ultrasonic distance measurement, wireless communication and intelligentized track movement control are combined together in an organic mode, the functions of precision location, automatic deviation rectification, track control, automatic anticollision and the like are achieved, and the purposes of lawn beautifying, energy conservation and environmental friendliness are achieved. The intelligentized precision automatic hay mower and the control device thereof are ingenious in structure, low in production cost, friendly in interface, convenient to operate, capable of saving energy, environmentally friendly and suitable for conducting finishing mowing on land parcels of medium-sized lawn or small-sized lawn.

A 3-dimensional computer generated image is generated by subdividing the image into a plurality of rectangular areas. Object data for each rectangular area is loaded into a display list memory (4) until that memory is substantially full. Image data and shading data for each picture element of each rectangular area is derived by an image synthesis processor (6) from the object data. Image data is then stored in a local memory (16) and further object data loaded into the display list memory (4) and replaces the existing contents. Once this has happened, the stored image data and shading data is retrieved and additional image data and shading data derived for each picture element by the image synthesis processor (6) using the new object data and the previously derived image and shading data. When there is no further object data to load to the display list memory the shading data is provided for display for the rectangular areas by a frame buffer 11. The memory may be managed by allocating at least one block of storage from the display list memory to each rectangular area and then storing in that block of memory, data which pertains to surfaces which intersect that rectangular area. A determination is made as to when a predetermined number of blocks have been used for a rectangular area. When this number is reached, the system starts to derive shading data for the rectangular areas, thereby releasing blocks of storage to be allocated to further rectangular areas in the display list memory.

The invention relates to a method for publishing an intelligent advertisement, which comprises the following steps: collection of the characteristic information of a user or a special user and the information of an area in which the user is located: inputting a picture or the video of a scene, scanning the picture or the video, recognizing the picture or the video and extracting the characteristic information of the user; comparing the characteristic information of the collected user or the special user with the characteristic information of a prestored user and judging whether the characteristic information of the collected user or the special user is matched with the characteristic information of a prestored user or not; searching for an advertisement in an advertisement resource library, which is matched with the characteristic information of the user; and outputting the advertisement. A system of the invention comprises an intelligent advertisement publishing unit or an array intelligent advertisement publishing unit managed by an integrated control center through an internetwork. The invention detects audiences by introducing technologies of scene scanning, computer image analysis, mode recognition, and the like, extracts the characteristic information to analyze, selects the advertisement which is matched with the characteristics of the audiences and presents the advertisement, thereby improving the effect of publishing the advertisement.

The present invention belongs to the computer image processing technical field and relates to a generative adversarial network model-based hairstyle changing method. There are a plenty of practical applications about automatic hairstyle changing in the classification and image editing field. According to the method of the present invention, a new large hairstyle data set HAIRSTYLE30k is provided, wherein the new large hairstyle data set HAIRSTYLE30k contains hairstyles composed of 64 different types of 30K images; and a hairstyle automatic generation and modification model H-GAN is provided to realize hairstyle automatic change. According to the method of the invention, improvements are made on the basis of the basic generative adversarial network model, and therefore, new data sets can be efficiently learned. The method not only performs well on a basic data set but also has good generalization on a new data set.

The invention relates to the technical field of computer images, in particular to a shooting training method and a shooting training system by using an augmented reality technology. The method comprises the following steps that: a real scene and a virtual shooting target are fused in real time by using the augmented reality technology, a battle scene is formed in video glasses of a trainer, a camera and a trigger on a gun system transmit shooting information to an intelligent mobile phone, the intelligent mobile phone judges whether the target is shot or not, and a shooting result is formed, stored and displayed in the video glasses worn on the head of the trainer; and meanwhile, the virtual target can shoot a shooter, a shooting result is formed, stored and displayed in the video glasses worn on the head of the trainer, and feedback effect information is output. The method and the system have the advantages that shooting training flexibility and reality simulation degree are high, and training cost is greatly saved.

There are provided methods and apparatus for generating a 3-dimensional computer image. The image comprises a number of objects and is divided into separate areas. In a first aspect, control data to link to object data stored in a memory for each object is derived for two objects at a time. This improves processing and memory usage. In a second aspect, two or more separate areas can be processed in parallel by deriving control data for the two separate areas at a time. This improves processing time. To avoid fetching data for both areas, that is actually only applicable to one area, encoding is used in the control data. In a third aspect, object data can be stored on one or across two memory pages, and the control data includes one memory page address in the former case and two memory page addresses in the latter case. This improves memory usage. In a fourth aspect, object data can be stored across two non-contiguous memory pages, by using a look-up table with contiguous portions allocated for each object's object data. This also improves memory usage. In a fifth aspect, when partial rendering is used, and partial depth data is stored and subsequently updated, the depth data can be stored across two memory pages using link portions in the memory pages to either indicate the end of data storage or point to the next memory page. This also improves memory usage.

The related optical detector for glass defect comprises: expanding the laser beam into plane beam to incidence from the target glass side; with the defect in glass as scatterer, such as air bubble and scratch, shooting the target on carrying platform by a camera controlled by computer to obtain the clear defect image to process by computer image process and recognition software, and displaying the defect mark and prompt. If adding an optical scanning mechanism included a rotating mirror and f-theta lens between the laser and column mirror, it can also be used for defect depth. This invention is simple and practicable for different target.

The invention relates to a method for automatically monitoring roadway deformation in real time. The method comprises the following steps of: distributing cross section marks on a selected cross section of a tested roadway section and distributing a plurality of observation mark points on the cross section; erecting a mining camera probe towards a monitored roadway in a mode that a main optical axis is vertical to the tested cross section; observing the initial true distance between previously-measured observation points, wherein the initial true distance is displayed as a display distance on a computer image; and acquiring a video image of the tested roadway section, monitoring the change of the image on monitoring center software of a computer in real time, processing the video image and automatically recognizing the cross section marks and the observation mark points so as to obtain a relative deformation value and an absolute deformation value in different statistics time scales or calculate the size and the change of the roadway cross section. The method has the advantages of automatic, real-time, continuous and non-contact monitoring, safety and high efficiency. A monitoring result can provide bases for the design of a roadway support, roadway maintenance and the prevention of rock burst and a roof disaster.

The invention discloses a face recognition method based on PCA image reconstruction and LDA and belongs to the technical field of computer image processing and pattern recognition. The face recognition method is based on a principal component analysis algorithm; an intra-class covariance matrix serves as a generation matrix for acquiring face feature subspaces of individuals; then an image to be recognized maps the feature subspaces to extract features; the image is reconstructed according to feature values; a residual image is subjected to the linear discriminant analysis; and finally, face recognition is realized by a minimum distance classification recognition algorithm. Compared with the prior feature subspace method, the face recognition method can better extract the face features of different people, and the face recognition rate is increased effectively. In addition, when a face database is required to be expanded, only the newly added faces are required for feature face training; not all the face feature subspaces are retrained; and the face recognition method also has good extendibility.

To provide a conference system, a monitoring system, an image processing apparatus, an image processing method and A non-transitory computer-readable storage medium that stores a computer-image processing program capable of accurately and effectively recognizing an object based on a fisheye-distorted image photographed at a wide angle.

When an instruction of selecting an arbitrary point Ci(ui, vi) of a 2D-viewable planar regular image generated based on a fisheye-distorted image S photographed by a fisheye lens is received, point Si(xi, yi) on the fisheye-distorted image S is calculated as a coordinate corresponding to Ci(ui, vi), and a pixel information group configuring the planar regular image is newly generated based on a pixel information group configuring the fisheye-distorted image S.

The invention relates to a detection and grading system for tenderness of fresh beef based on multispectral imagery, which belongs to the technical field of detection of food quality. In a dark box, the lens of a CCD industrial camera is aligned with the center of a table board of a height-adjustable objective table, two multispectral light source assemblies are symmetrically arranged at a left sidewall and a right sidewall of the dark box respectively through two light source fixing supports, and a large power LED array on each of multispectral light source assemblies is composed of LEDs of 6 kinds of wave bands; during detection, a sample to be detected is put on the table board of the height-adjustable objective table, the LEDs of 6 kinds of wave bands alternately radiate in order, the CCD industrial camera acquires multispectral images of the sample to be detected, a multispectral beef sample reflection image in an interested area of a lean meat part is extracted through computer image processing, and therefore, marbling of beef is extracted, and evaluation and grading of content of marbling of the beef are carried out, and tenderness of the beef is predicted. The detection and grading system provided by the invention is applicable to companies dealing with cutting, marketing, processing and import and export of fresh beef and to market supervision by quality and technical supervision departments.