11 results about "Image sequence" patented technology

The invention discloses a real-time gesture recognition method which comprises the steps of (1) decomposing an obtained gesture video into image sequences sorted in a chronological order and preprocessing obtained images and then carrying out hand-region segmentation, (2) extracting a hand shape feature of a hand region in each image and using an SVM support vector machine to identify the hand shape feature as a corresponding gesture value, (3) combining the gesture value of each image and a direction feature of a motion trajectory obtained by an iterating LK pyramid optical flow algorithm asa feature vector of each dynamic gesture image, (4) carrying out loop execution of steps (2) and (3) with a loop end condition that all images of a current video are processed so as to obtain a complete set of feature vector sequences, (5) creating a gesture template library, (6) carrying out optimization DTW match on the obtained feature vector sequences and all templates in a template library, calculating the degree of distortion of the match, wherein recognition is failed if the degree of distortion is larger than a distortion threshold, and a recognition result is outputted if the degree of distortion is smaller than the distortion threshold.

A method for enhancing detection of a moving object. The method includes comparing a video image to a historical sequence of images obtained from the same scene or stationary source. The method uses statistical sampling to implement change detection and thresholding operations which extract moving objects from the scene. The method provides for a high degree of sensitivity and noise rejection to motion detection, and is self-adapting to variations in sensor characteristics and operating environments.

The invention relates to a self-adaptive optical image high resolution restoration method combined with frame selection and blind deconvolution, comprising the following steps: firstly, a short exposure image sequence gn (x, y) is recorded when a self-adaptive optical closed loop is corrected; shannnon entropy of each frame of image in the sequence is calculated; a degraded image gm (x, y) with lower entropy is selected for blind deconvolution image restoration; secondly, an initial value hm (x, y) of a point spread function is generated by utilization of random phase; thirdly, a target f(x, y) is estimated by using the gm(x, y) and the obtained hm (x, y), and an estimated value f(x, y) is obtained after addition of positivity limitation on the target; fourthly, an estimated value hm (x, y) of the point spread function is obtained by using the gm (x, y) and the f(x, y), and an estimated value h(x, y) is obtained after addition of positivity limitation in the same way; fifthly, inspection is made whether an iterated value h(x, y) and an iterated value f(x, y) meet iteration stopping requirements or not; if the iterated values do not meet the iteration stopping requirements, the third step is returned; if the iterated values meet the iteration stopping requirements, circulation is stopped and the f(x, y) and the h(x, y) are outputted. The invention has the advantages of effective improvement of restoration quality, acceleration of convergence rate, capability of well compensating correction capability under hardware limitation of a self-adaptive optical system, and improvement of imaging quality.

The invention discloses a method for GPU-based (Graphics Processing Unit) object three-dimensional shape measurement, and belongs to the technical field of object three-dimensional measurement. The method particularly comprises the following steps: projecting structure grating to the surface of an object to be measured by a projector; shooting the surface of the object to be measured by two cameras simultaneously to obtain two groups of image sequences; storing the two groups of image sequences into computer memory; completing distortion correction, phase demodulation, corresponding points matching and three-dimensional reconstruction for the image by a GPU in parallel; transmitting the object three-dimensional coordinate obtained through three-dimensional reconstruction to the computer memory by the GPU; repeating the above steps until the measurement is complete, and releasing the GPU and the storing space. The method adopts multi-core GPU to realize the parallel calculation at thread level, greatly increases data processing speed, and can satisfy the real-time requirement for three-dimensional measurement of a dynamic object.

The invention relates to a monocular texture-free three-dimensional object attitude tracking method based on a three-dimensional geometric model. The method of the invention uses the edge features ofthe object to carry out attitude estimation, does not depend on the feature points of the object surface, and is suitable for weak texture or texture-free three-dimensional object tracking. The methodof the invention can solve the attitude parameters of a texture-free three-dimensional object (the rotation and translation parameters of the object relative to the camera) in real time without installing a mark or a hardware positioning sensor on the three-dimensional object, the virtual object generated by computer is aligned with the three-dimensional object in the image sequence, and the virtual-real superposition effect of seamless fusion is generated.

The invention relates to a subcutaneous microvessel segmentation and three-dimensional reconstruction method based on optical coherence tomography. The subcutaneous microvessel segmentation and three-dimensional reconstruction method comprises the following steps of obtaining a two-dimensional image sequence through an FFOCT system and preprocessing; grouping the acquired two-dimensional images, and performing robust principal component analysis on each group of images to obtain an image background and a high-response microvascular region; and carrying out capillary segmentation and three-dimensional reconstruction on the image background and the high-response capillary region in combination with the position relationship between each group of images and the capillary boundary distributionin each group of images. Compared with the prior art, the method has the advantages that the global optical coherence tomography technology is applied to image recognition of venipuncture, the imaging precision is remarkably improved, blood vessel positioning is more accurate, and capillary clear imaging is more reliable.