70 results about "Spin image" patented technology

A new measure to compare histogram-based descriptors, a diffusion distance, is disclosed. The difference between two histograms is defined to be a temperature field. The relationship between histogram similarity and diffusion process is discussed and it is shown how the diffusion handles deformation as well as quantization effects. As a result, the diffusion distance is derived as the sum of dissimilarities over scales. Being a cross-bin histogram distance, the diffusion distance is robust to deformation, lighting change and noise in histogram-based local descriptors. In addition, it enjoys linear computational complexity which significantly improves previously proposed cross-bin distances with quadratic complexity or higher The proposed approach is tested on both shape recognition and interest point matching tasks using several multi-dimensional histogram-based descriptors including shape context, SIFT and spin images. In all experiments, the diffusion distance performs excellently in both accuracy and efficiency in comparison with other state-of-the-art distance measures. In particular, it performs as accurate as the Earth Mover's Distance with a much greater efficiency.

The invention discloses an image rotation correction method and a system and an electronic device. The image rotation correction method carries out the following steps before storing or browsing images: A: reading original image data; B. detecting the rotation angle of the image data: rotating images several times, calculating the reliability indices of the image before rotation and the images after each rotation, and marking the rotation angle whose reliability index is highest as the needed rotation angle, wherein, the rotation angle is the relative angle of the original image and image during normal checking; and C. according to detected rotation angle, carrying out rotation correction on the image. The invention utilizes the technologies in the fields of image processing, mode identification and the like to automatically detect the rotation angle and correct the rotation of the image data before storing or displaying the image data to a user to browse, thus ensuring that the user can see the image which is treated by rotation correction when browsing images and reduces the trouble for users to manually correct with an image browsing tool.

A contact developing method including supplying a two-component developer to an electrostatic latent image on a rotating image bearing member by rotating a developing sleeve and a rotatable magnet having multiple magnetic poles provided inside the developing sleeve, to develop the electrostatic latent image into a toner image. The developing sleeve and the image bearing member rotate in the same direction while facing each other. The two-component developer comprises a non-magnetic toner and a carrier. The carrier comprises a magnetic core particle and a resin layer covering the magnetic core particle. The resin layer comprises a conductive particle and a resin. The conductive particle comprises an alumina-based material and a conductive layer covering the alumina-based material. The resin is obtained by heating a copolymer comprising a monomer A unit and a monomer B unit.

A parallax image pickup apparatus for forming a parallax image train by photographing an object from a plurality of directions includes an image pickup unit in which an image pickup device and an image forming optical system are integrated in order to photograph the parallax images of the object; a rotating unit on which the image pickup unit is mounted and which rotates the image pickup unit in a parallax direction; an image pickup unit moving unit for rectilinearly moving the image pickup unit and the rotating unit synchronously with the rotation of the rotating unit; a control unit for controlling so that a virtual image pickup center portion is always located at the center of a photographed image; and a focal distance control unit for continuously changing a focal distance of the image pickup unit.

Methods and apparatus for performing an inline rotation of an image. The apparatus includes a rotation unit for reading pixels from a source image in an order based on a specified rotation to be performed. The source image is partitioned into multiple tiles, the tiles are processed based on where they will be located within the rotated image, and each tile is stored in a tile buffer. The target pixel addresses within a tile buffer are calculated and stored in a lookup table, and when the pixels are retrieved from the source image by the rotation unit, the lookup table is read to determine where to write the pixels within a corresponding tile buffer.

The method of eliminating image rotation caused by 45 deg scanning lens is suitable for eliminating image rotatino in case of faulted K lens and no K lens. The method includes at least the folloiwng steps: rotating computer input image; calculating imaging relationship in computer; deducing the imaging process of the optical instrument from the basic reflection law; obtaining the mapping relationship between the image rotating coordinate and rotation elimianting image coordinate; eliminating rotation of the whole image frame; and output rotation eliminated image; circular execution. For faulted K lens, there include also the steps of obtaining K lens angle and inputting to computer. The present invention can eliminate rotation of image, simplify optical path of remote sensing instrument, raise the reliability of instrument in satellite.

The invention provides a dual-mode image automatic fusion method which comprises the following steps: respectively obtaining distance sequences from IVUS and OCT (Optical Coherence Tomography) inner membrane points to the center of an image on the basis of a sinake model; Calculating scaling coefficients of the OCT image and the IVUS image blood vessel lumen through the average inner membrane radius of the IVUS image and the average inner membrane radius of the OCT image blood vessel lumen; Rotating the IVUS image according to a preset angle range and a preset step length; calculating mutual information of the rotated IVUS image and the OCT image; when the mutual information is the maximum value, the corresponding angle serves as a reference, adopting a circumferential correlation algorithm for estimating the actual rotation angle, and after IVUS and OCT images are registered, adoptinga Poisson fusion algorithm for fusing the converted IVUS image and the converted OCT original image. According to the technical scheme, the advantages of IVUS imaging and OCT imaging are brought into full play, and the speed of the fusion process is increased.

An Image re-sampling tampering detection algorithm is one of the important methods for image tampering evidence collection. According to the invention, an image is detected mainly based on condition state transition probability matrixes in a Markov process and the texture features in a Gabor filter. After an image is re-sampled, the correlation between adjacent DCT coefficients changes. However, the transition probability matrixes can well describe the change in DCT coefficients and surrounding coefficients. Thus, transition probability matrixes in four directions are extracted as Markov features. The detection rate of the algorithm fluctuates greatly when comes to texture images of different complexity. Thus, a Gabor filter group is adopted to extract the texture features of the image and perform two-dimensional discrete Fourier transform to get the final texture features of the image, and then, an SVM is adopted to train and classify the two types of features. With the algorithm of the invention, scalable images and rotary images can be well detected, and detection of rotary images is of high robustness.

The invention discloses an intracranial aneurysm detection method based on MRA, and the method comprises the following steps: obtaining to-be-processed original image data, carrying out the correction and normalization processing of the original image data, and obtaining the processed image data; performing mIP reconstruction on the processed image data to obtain MIP images, and generating three kinds of rotating images after the external carotid artery is deleted, wherein each rotating image is composed of projections evenly distributed around a single rotating shaft; automatically segmenting intracranial artery blood vessel voxels; manufacturing a plurality of patches for each voxel in the blood vessel area, calculating the probability that all patches contain the aneurysm, classifying the calculated probabilities according to a preset probability threshold value, wherein the voxels larger than the preset probability threshold value generate a volume shape index image and a curvature image based on a curvature algorithm to display the aneurysm; measuring the maximum diameter of the aneurysm; and outputting an aneurysm detection result. According to the method, the intracranial artery blood vessel can be accurately identified, and the size of the aneurysm can be accurately measured.

The scanning type printing apparatus prints an image and/or a text on a recording medium in accordance with image data and/or text data while moving on the recording medium. The apparatus includes a moving unit, an image processing unit that, prior to printing, rotates the image and/or the text to be printed based on the image data and/or the text data and/or reverses front and back thereof; and an ejection head that includes at least one nozzle and at least one ink ejection unit that eject ink.

Methods, systems, computer-readable media, and apparatuses are presented for generating a photorealistic viewable model using augmented reality (AR). An AR scene is generated by overlaying a virtual object onto a view of a physical environment. When placed into the AR scene, the virtual object can interact with the physical environment by, for example, reflecting or taking on colors, shadows, brightness, and other attributes of the physical environment. To generate the viewable model, the virtual object is manipulated (e.g., moved or rotated) within the AR scene and a plurality of images are generated by capturing the virtual object as the virtual object is being manipulated. The viewable model can be generated based on one or more of the images and can be output in the form of an interactive presentation, for example, a spin image.

The invention discloses a typhoon positioning method, device and system, electronic equipment and a storage medium. The typhoon positioning method comprises the following steps of: acquiring previous moment geographic position information and a previous moment satellite image of a typhoon; determining first pixel coordinates of the previous moment geographic position information in the previous moment satellite image; taking the first pixel coordinates as a center, taking a target image area with a set window size from the previous moment satellite image data, and rotating the target image area to obtain a plurality of rotating image areas; respectively matching each rotating image area with a current moment satellite image, and determining the central point of the rotating image area with the highest matching degree as a current moment first typhoon position; and converting the current moment first typhoon position into current moment geographical position information. According to the typhoon positioning method, multiple data sources are adopted, the problem that errors are large when typhoon positioning is conducted only through satellite images is avoided through adopting a multi-source weighted fusion algorithm and a distance attenuation algorithm, and the calculated typhoon path is smoother and more stable.

The invention provides a light spot detection method and device, electronic equipment and a readable storage medium, and the method comprises the steps: carrying out the scanning of an original image and a rotating image obtained after the original image rotates 90 degrees in a first direction in a horizontal direction, obtaining the pixel points of each row of images in the original image and the rotating image, determining candidate light spot pixel points from the pixel points, forming a row light spot array by all candidate light spot pixel points meeting a preset requirement in each row image, and respectively clustering all the row light spot arrays to obtain target light spot arrays of the original image and the rotating image, and performing union set on the spot area corresponding to the target light spot array of the rotating image after rotating for 90 degrees along the second direction and the spot area corresponding to the target light spot array of the original image to obtain a target light spot of the original image. According to the invention, light spot detection can be carried out on the detection image simultaneously through one set of algorithm, the problem of high memory consumption is solved, and the effects of simple algorithm, small storage space, low time consumption and accurate and stable light spot center point extraction are achieved.

The invention relates to the field of multimedia communication, in particular to a video call processing method and device, and the method comprises the steps: obtaining the angle information of a current transmitting end in a video call process; rotating a currently acquired image according to the angle information and a preset image acquisition angle; displaying rotated images, the rotated imageis sent to a receiving end; and the receiving end displays the rotated image. Therefore, according to the change of the receiving end angle information; the image is rotated, the rotated image is displayed locally and sent to the receiving end, and therefore it can be guaranteed that the locally-displayed video image and the video image displayed by the receiving end are both upright normal images, image rotation is conducted through the receiving end, the processing efficiency is high, real-time synchronism can also be guaranteed, and the accuracy is improved.

The invention discloses a rotating palm image detection method, which comprises the following steps of: 1, rotating an image to be detected at an interval of 5 degrees to form a 72-original image, andstoring the 72-original image into an image rotating star disk from a rotating image within a range of 0-360 degrees; 2, inputting all the pictures in the image rotation star disk into a pre-trainedforward palm detector in batches according to a sequence; 3, screening all detection results to the maximum extent, and obtaining a unique candidate detection result; 4, calculating a reverse rotationmatrix corresponding to the rotation angle in the candidate detection result; and finally, performing rotation processing on the candidate detection result through the rotation matrix to obtain a final palm image detection result with an angle. The invention provides an image rotation star disc and a maximum possibility screening method, and combines the most efficient detection network Yolov3 inthe target detection field in the current deep learning to realize palm image detection with an attached angle.

Systems and methods are provided for image re-orientation. A computing system for image re-orientation may comprise one or more processors and a memory for storing instruction that, when executed by the processors, cause the system to perform: obtaining positive samples including text of a predetermined type in a predetermined orientation range and negative samples not including text of the predetermined type in the predetermined orientation range; training a model with the obtained positive and negative samples; receiving an image in an original orientation; applying the trained model to thereceived image; in response to determining that the received image does not comprise any text of the predetermined type in the predetermined orientation range, rotating the image by a predetermined angle and re-applying the trained model to the rotated image to determine if the rotated image comprises at least some text of the predetermined type in the predetermined orientation range.