The keyframe image character of 3D animation refers to the image state of the character or scene and other objects presented at specific keyframes in the animation timeline. The character texture will be blurred, dist...
详细信息
The keyframe image character of 3D animation refers to the image state of the character or scene and other objects presented at specific keyframes in the animation timeline. The character texture will be blurred, distorted and discontinuous in the animation sequence due to motion, perspective change or lighting difference. Due to the lack of optical flow information to guide the current 3D animation keyframe image enhancement process, it is difficult to effectively distinguish the foreground character from the background elements, which affects the algorithm's ability of recognizing the animation character's movement patterns. To this end, a character texture enhancement method based on the sift algorithm and LK (Lucas-Kanade) optical flow is investigated for 3D animation keyframe images. The sift algorithm is used to extract the feature points of the character in each frame of 3D animation image as the initial tracking template, the LK optical flow method is used to calculate the optical flow vector of the feature points in the new frame, and the weighted directional histogram is used to estimate the direction of the character's movement, and then the feature points in the tracking template are optimized through feature matching and weight updating to complete the tracking of the character's target in each frame of the image. According to the target tracking results, a fractional-order differential mask is constructed in the character target region, and an adaptive function is constructed by combining the modes of the local gradient, information entropy and variance parameters of the image, so as to realize the enhancement of the character texture of the key frame images of the 3D animation by dynamically adjusting the order of the fractional-order differential. The experimental results show that the method can realize the dynamic tracking of 3D animation image character, and the extracted 3D animation image character feature points show a balanced distribution, an
In river management, it is important to obtain ice velocity quickly and accurately during ice flood periods. However, traditional ice velocity monitoring methods require buoys, which are costly and inefficient to dist...
详细信息
In river management, it is important to obtain ice velocity quickly and accurately during ice flood periods. However, traditional ice velocity monitoring methods require buoys, which are costly and inefficient to distribute. It was found that UAV remote sensing images combined with machine vision technology yielded obvious practical advantages in ice velocity monitoring. Current research has mainly monitored sea ice velocity through GPS or satellite remote sensing technology, with few reports available on river ice velocity monitoring. Moreover, traditional river ice velocity monitoring methods are subjective. To solve the problems of existing time-consuming and inaccurate ice velocity monitoring methods, a new ice velocity extraction method based on UAV remote sensing technology is proposed in this article. In this study, the Mohe River section in Heilongjiang Province was chosen as the research area. High-resolution orthoimages were obtained with a UAV during the ice flood period, and feature points in drift ice images were then extracted with the scale-invariant feature transform (sift) algorithm. Moreover, the extracted feature points were matched with the brute force (BF) algorithm. According to optimization results obtained with the random sample consensus (RANSAC) algorithm, the motion trajectories of these feature points were tracked, and an ice displacement rate field was finally established. The results indicated that the average ice velocities in the research area reached 2.00 and 0.74 m/s, and the maximum ice velocities on the right side of the river center were 2.65 and 1.04 m/s at 16:00 on 25 April 2021 and 8:00 on 26 April 2021, respectively. The ice velocity decreased from the river center toward the river banks. The proposed ice velocity monitoring technique and reported data in this study could provide an effective reference for the prediction of ice flood disasters.
The traditional image recognition technology can transform some expression form of image into the data which can be processed by computer, and recognize the image with decision function. However, in actual application...
详细信息
The traditional image recognition technology can transform some expression form of image into the data which can be processed by computer, and recognize the image with decision function. However, in actual applications, incomplete 3D images will be encountered. In order to screen the required image information from a large amount of images, it is necessary to recognize and match the image, and so the research has long-term application value. In this paper, sift algorithm was used to extract the feature vectors for incomplete 3D information recognition. Then, the modeling method of circular matching pattern was proposed, and the pattern of mathematical recognition was adopted in the process of image recognition. After that, on the basis of a large number of domestic and foreign research literatures, the process of "image hypothesis" converted to "image recognition" was completed. Finally, the system simulation was carried out by using Microsoft Visual Studio. For the incomplete 3D information image, the image feature recognition system was completed.
The initial assessment of any skin disease is usually made by visual inspection of doctors and skin specialists. Further tests may be recommended such as biopsy and pathological examination for a more accurate diagnos...
详细信息
ISBN:
(纸本)9781510638464
The initial assessment of any skin disease is usually made by visual inspection of doctors and skin specialists. Further tests may be recommended such as biopsy and pathological examination for a more accurate diagnosis. With the use of skin disease identification system, the diagnosis of infected skin is readily attainable without undergoing biopsy and pathological examination. The infected skin disease image is identified using sift algorithms with local features and K-NN classifier. The skin disease that will be identified are acne, psoriasis, eczema, rashes, hives, warts, tinea versicolor (an-an) and unknown skin disease. The system was confirmed to be efficient in identifying the aforementioned skin diseases. Identification of infected skin images is accomplished by K-Nearest Neighbors (K-NN) algorithm which shows an accuracy of 90% in functionality testing.
In this paper, face liveness detection method which is based on very short video is proposed. Firstly, spatial and temporal video processing is performed by using Eulerian video magnification. Next, the matching pairs...
详细信息
ISBN:
(纸本)9781728165905
In this paper, face liveness detection method which is based on very short video is proposed. Firstly, spatial and temporal video processing is performed by using Eulerian video magnification. Next, the matching pairs between frames are extracted from the original video. Then, the paper constructs the feature histogram from the color difference between two pixels according to sift matching pairs. Finally, we use SVM classifier to determine the liveness of the subject. The experimental results show that the detection accuracy is 95% and the liveness detection can be completed in less than one second without any specific user action.
Image mosaic technology has developed rapidly in recent years. The core of image mosaic is the processing method of image registration. sift algorithm has become a common method of image registration because of its si...
详细信息
Image mosaic technology has developed rapidly in recent years. The core of image mosaic is the processing method of image registration. sift algorithm has become a common method of image registration because of its simple and accurate characteristics. This paper reviews a large number of literatures and summarizes the general process of sift algorithm, which includes scale space construction, key point detection and elimination, key point direction assignment, key point description and key point matching. After these processing, the target image can be fused and reconstructed into a new image with a wide view angle and more information.
Considering the low real-time performance and the large amount of false matches exist in the feature matching stage of traditional Scale Invariant Feature Transform(sift) algorithm in unmanned aerial vehicle (UAV) rem...
详细信息
ISBN:
(纸本)9781538691540
Considering the low real-time performance and the large amount of false matches exist in the feature matching stage of traditional Scale Invariant Feature Transform(sift) algorithm in unmanned aerial vehicle (UAV) remote sensing image registration. In this paper, a series of optimization methods for traditional sift algorithms are proposed, including sift execution process optimization, changing the parameters of scale space construction, Simplified method for judging the construction area of feature descriptor and construction of bidirectional matching filters. Experiments on UAV remote sensing images show that the optimization method can significantly improve the matching efficiency compared with traditional methods, and the comprehensive acceleration ratio is about 35% to 40%, which proves the effectiveness of the acceleration method.
The train uncoupling robot is a 4-DOF robot which can complete the uncoupling work efficiently and accurately in the process of disassembly or marshalling of freight trains or passenger trains without the help of work...
详细信息
ISBN:
(纸本)9781728101057
The train uncoupling robot is a 4-DOF robot which can complete the uncoupling work efficiently and accurately in the process of disassembly or marshalling of freight trains or passenger trains without the help of workers. The uncoupling work should be completed when the train and the uncoupling robot are both in motion. However, this increases the difficulty of recognizing the hook of trains and finishing the uncoupling work automatically. Therefore, the visual system of the robot plays a vital role during the uncoupling process so that it can be able to recognize the target hook by analyzing images and data from cameras and sensors. Furthermore, it should also have the ability to predict the motion planning of the object. In this paper, a visual control scheme is developed by using the Scale-invariant Feature Transform (sift) algorithm to recognize the hook of train in motion. It searches the extreme points as the key-points, and extracts the features such as position, scale and direction to match between the two images. The results indicate that the target hook can be recognized accurately and the recognition speed can meet the working requirements of the system by using the proposed visual control scheme even if the size of the target is varying or the target rotates moderately.
Coin recognition is one of the prime important activities for modern banking and currency processing systems in which machine vision is widely used. The technique at the heart of such systems is object recognition in ...
详细信息
ISBN:
(纸本)9781538630136
Coin recognition is one of the prime important activities for modern banking and currency processing systems in which machine vision is widely used. The technique at the heart of such systems is object recognition in a digital image. Although it has high recognition speed, the traditional method of coin recognition can not recognize the coins with similar sizes. This paper presents a method based on sift(scale invariant feature transform) algorithm for coin recognition. sift algorithm can handle the issues of rotations, scaling and illumination in a digital image. Therefore it can solve the problem about distinguishing the coins which have approximate size. In experiments, we compare the performances of Chinese coin recognition with our proposed method and the traditional method(based on size). The results demonstrate the feasibility and effectiveness of our approach.
In this paper, we propose an image registration algorithm based on improved sift (Scale-Invariant Feature Transform) algorithm and essential matrix estimation based on RANSAC (Random Sample Consensus) and AC-RANSAC (A...
详细信息
ISBN:
(纸本)9781538631546
In this paper, we propose an image registration algorithm based on improved sift (Scale-Invariant Feature Transform) algorithm and essential matrix estimation based on RANSAC (Random Sample Consensus) and AC-RANSAC (A Contrario RANSAC) algorithm. So that in the 3D reconstruction, we can directly restore the parameters of the camera by using the essential matrix model estimated by image registration algorithm. The essential matrix is a 5-parameter model, reflecting the relationship between the representation of the spatial image points in the camera coordinate system under different viewing angles. sift algorithm not only maintains the invariance of scale, rotation, brightness and so on, but also maintains a certain degree of stability to the angle change, affine transformation and noise, but the time performance is low and the matching accuracy is not high enough. Therefore, we propose to narrow the dimension of the sift feature vector to reduce the time consumption, and increase the similarity measure of the nearest neighbor distance less than 0.3 to calculate the feature point correspondence between images. The experimental results have demonstrated that our method not only can guarantee better time performance, but also can effectively eliminate the wrong match point, greatly improving the matching accuracy.
暂无评论