Abstract: Traditional museums and most digital museums use window display to exhibit their collections. However, the agricultural tools are distinctive for their use value and wisdom contained. Therefore, this paper first proposes a method of virtual interactive display for agricultural tools based on virtual reality technology, which combines static display and dynamic use of agricultural tools vividly showing the agricultural tools. To address the problems of rigid interaction and terrible experience in the process of human-computer interaction, four human-computer interaction technologies are proposed to design and construct a more humanized system including intelligent scenes switching technology, multichannel introduction technology, interactive virtual roaming technology, and task-based interactive technology. The evaluation results demonstrate that the system proposed achieves good performance in fluency, instructiveness, amusement, and practicability. This human-computer interaction system can not only show the wisdom of Chinese traditional agricultural tools to the experiencer all over the world but also put forward a new method of digital museum design. PubDate: Tue, 01 Jan 2019 00:00:00 +000
Abstract: Perceptual hashing technique for tamper detection has been intensively investigated owing to the speed and memory efficiency. Recent researches have shown that leveraging supervised information could lead to learn a high-quality hashing code. However, most existing methods generate hashing code by treating each region equally while ignoring the different perceptual saliency relating to the semantic information. We argue that the integrity for salient objects is more critical and important to be verified, since the semantic content is highly connected to them. In this paper, we propose a Multi-View Semi-supervised Hashing algorithm with Perceptual Saliency (MV-SHPS), which explores supervised information and multiple features into hashing learning simultaneously. Our method calculates the image hashing distance by taking into account the perceptual saliency rather than directly considering the distance value between total images. Extensive experiments on benchmark datasets have validated the effectiveness of our proposed method. PubDate: Mon, 19 Nov 2018 00:00:00 +000
Abstract: Video segmentation into shots is the first step for video indexing and searching. Videos shots are mostly very small in duration and do not give meaningful insight of the visual contents. However, grouping of shots based on similar visual contents gives a better understanding of the video scene; grouping of similar shots is known as scene boundary detection or video segmentation into scenes. In this paper, we propose a model for video segmentation into visual scenes using bag of visual word (BoVW) model. Initially, the video is divided into the shots which are later represented by a set of key frames. Key frames are further represented by BoVW feature vectors which are quite short and compact compared to classical BoVW model implementations. Two variations of BoVW model are used: classical BoVW model and Vector of Linearly Aggregated Descriptors (VLAD) which is an extension of classical BoVW model. The similarity of the shots is computed by the distances between their key frames feature vectors within the sliding window of length , rather comparing each shot with very long lists of shots which has been previously practiced, and the value of is . Experiments on cinematic and drama videos show the effectiveness of our proposed framework. The BoVW is -dimensional vector and VLAD is only -dimensional vector in the proposed model. The BoVW achieves segmentation accuracy, whereas VLAD achieves . PubDate: Thu, 08 Nov 2018 08:36:58 +000
Abstract: Visualization provides an interactive investigation of details of interest and improves understanding the implicit information. There is a strong need today for the acquisition of high quality visualization result for various fields, such as biomedical or other scientific field. Quality of biomedical volume data is often impacted by partial effect, noisy, and bias seriously due to the CT (Computed Tomography) or MRI (Magnetic Resonance Imaging) devices, which may give rise to an extremely difficult task of specifying transfer function and thus generate poor visualized image. In this paper, firstly a nonlinear neural network based denoising in the preprocessing stage is provided to improve the quality of 3D volume data. Based on the improved data, a novel region space with depth based 2D histogram construction method is then proposed to identify boundaries between materials, which is helpful for designing the proper semiautomated transfer function. Finally, the volume rendering pipeline with ray-casting algorithm is implemented to visualize several biomedical datasets. The noise in the volume data is suppressed effectively and the boundary between materials can be differentiated clearly by the transfer function designed via the modified 2D histogram. PubDate: Thu, 01 Nov 2018 00:00:00 +000
Abstract: The height estimation of the target object is an important research direction in the field of computer vision. The three-dimensional reconstruction of structured light has the characteristics of high precision, noncontact, and simple structure and is widely used in military simulation and cultural heritage protection. In this paper, the height of the target object is estimated by using the word structure light. According to the height dictionary, the height under the offset is estimated by the movement of the structured light to the object. In addition, by effectively preprocessing the captured structured light images, such as expansion, seeking skeleton, and other operations, the flexibility of estimating the height of different objects by structured light is increased, and the height of the target object can be estimated more accurately. PubDate: Thu, 01 Nov 2018 00:00:00 +000
Abstract: The study of the self-protective behaviors of dairy cows suffering dipteral insect infestation is important for evaluating the breeding environment and cows’ selective breeding. The current practices for measuring diary cows’ self-protective behaviors are mostly by human observation, which is not only tedious but also inefficient and inaccurate. In this paper, we develop an automatic monitoring system based on video analysis. First, an improved optical flow tracking algorithm based on Shi-Tomasi corner detection is presented. By combining the morphological features of head, leg, and tail movements, this method effectively reduces the number of Shi-Tomasi points, eliminates interference from background movement, reduces the computational complexity of the algorithm, and improves detection accuracy. The detection algorithm is used to calculate the number of tail, leg, and head movements by using an artificial neural network. The accuracy range of the tail and head reached [0.88, 1] and the recall rate was [0.87, 1]. The method proposed in this paper which provides objective measurements can help researchers to more effectively analyze dairy cows’ self-protective behaviors and the living environment in the process of dairy cow breeding and management. PubDate: Wed, 10 Oct 2018 00:00:00 +000
Abstract: Today, we face different approaches to enhance the robustness of image watermarking schemes. Some of them can be implemented, but others in spite of spending money, energy, and time for programming purpose would fail because of not having a strong feasibility study plan before implementation. In this paper, we try to show a rational feasibility study before implementation of an image watermarking scheme. We develop our feasibility study by proposing three types of theoretical, mathematical, and experimental deductions. Based on the theoretical deduction, it is concluded that the “S” coefficients in the second level of Singular Value Decomposition (SVD) offer high robustness to embed watermarks. To prove this, a mathematical deduction composed of two parts is presented and the same results were achieved. Finally, for experimental deduction, 60 different host images in both normal and medical images from various sizes of 256256 to 10241024 were imposed to 9 common geometric and signal processing attacks and the resistances of “S” coefficients against the attacks in the first and second levels of SVD were compared. Experimental result shows significant enhancement in stability and robustness of the “S” coefficients in the second level of SVD in comparison to the first level. Consequently all theoretical, mathematical, and experimental deductions confirmed domination of the “S” coefficients in the second level of SVD than the first level. In this paper, we do not show any specific implementation for the watermarking scheme. Instead, we investigate the potential performance gains from the singular values (S), of the second level of SVD and Discrete Wavelet Transform (DWT), and prove their superiority in comparison to conventional SVD+DWT watermarking schemes. PubDate: Wed, 10 Oct 2018 00:00:00 +000
Abstract: To discover the influence of the commercial videos’ low-level features on the popularity of the videos, the feature selection method should be used to get the video features influencing the videos’ evaluation mostly after analyzing the source data and the audiences’ evaluations of the videos. After extracting the low-level features of the videos, this paper improved the Correlation-Based Feature Selection (CFS) method which is widely used and proposed an algorithm named CFS-Spearmen which combined the Spearmen correlation coefficient and the classical CFS to select features. The 4 datasets in UCI machine learning database were employed as the experiment data. The experiment results were compared with the results using traditional CFS, Minimum Redundancy and Maximum Relevance (mRMR). The SVM was used to test the method in this paper. Finally, the proposed method was used in commercial videos’ feature selection and the most influential feature set was obtained. PubDate: Wed, 10 Oct 2018 00:00:00 +000
Abstract: With the rapid development of multimedia technology, the way of obtaining high-quality motion reproduction for space targets has attracted much attention in recent years. This paper proposes a Perception-driven Transcale Display Scheme, which significantly improves the awareness of multimedia processing. This new scheme contains two important modules, transcale description based on visual saliency and perception-driven display of space image sequences. The former concentrates on describing the transcle feature of space targets, including three algorithms, attention region computing, frame rate conversion, and image resolution resizing. On this basis, the latter focuses on high-quality display of space movements under different scales, including three algorithms, namely, target trajectory computing, space transcale display, and space movement display. Extensive quantitative and qualitative experimental evaluations demonstrate the effectiveness of the proposed scheme. PubDate: Thu, 04 Oct 2018 00:00:00 +000
Abstract: Vehicle detection and classification are very important for analysis of vehicle behavior in intelligent transportation system, urban computing, etc. In this paper, an approach based on convolutional neural networks (CNNs) has been applied for vehicle classification. In order to achieve a more accurate classification, we removed the unrelated background as much as possible based on a trained object detection model. In addition, an unsupervised pretraining approach has been introduced to better initialize CNNs parameters to enhance the classification performance. Through the data enhancement on manual labeled images, we got 2000 labeled images in each category of motorcycle, transporter, passenger, and others, with 1400 samples for training and 600 samples for testing. Then, we got 17395 unlabeled images for layer-wise unsupervised pretraining convolutional layers. A remarkable accuracy of 93.50% is obtained, demonstrating the high classification potential of our approach. PubDate: Tue, 02 Oct 2018 07:39:38 +000
Abstract: For massive crowds, users often have the need for interactive roaming. A good roaming effect can make the user feel immersed in the crowd, and the scenes need to be populated with crowds of people that make the environment both alive and believable. This paper proposes a method of efficiency control for massive crowd rendering. First, we devise a state machine mechanism based on self-feedback, which can dynamically adjust the accuracy of crowd model rendering according to the relationship between the speed of the system rendering and the speed the users expect. Second, we propose a movement frequency update method to perform the frequency of motion update based on the distance between the individual and the viewpoint. In addition, we propose a variable precision point sampling drawing strategy to render the individual with different sampling precision. The state machine system in this paper effectively integrates two core technologies for dynamically controlling the accuracy of the model, ensuring visual efficiency, improving the rendering efficiency, and satisfying the fluency of users’ roaming interaction. PubDate: Mon, 01 Oct 2018 00:00:00 +000
Abstract: The least significant bit (LSB) substitution techniques are detected by RS analysis and the traditional pixel value differencing (PVD) approaches are detected by pixel difference histogram (PDH) analysis. The PVD steganography can escape from PDH analysis by using the edges in multiple directions. This paper proposes a steganography technique by exploiting the edges in eight directions and also using LSB substitution to resist from both RS analysis and PDH analysis. For every 3×3 pixel block the central pixel is embedded with 3 or 4 bits of data by modified LSB substitution technique. Then this new value of the central pixel is utilized to calculate eight difference values with eight neighboring pixels. These eight difference values are used to hide the data. There are two types with regard to two different range tables. Type 1 uses 3 bit modified LSB substitution and range table 1. Type 2 uses 4 bit modified LSB substitution and range table 2. Type 1 and type 2 are also known as variant 1 and variant 2, respectively. Type 1 possesses higher PSNR and type 2 possesses higher hiding capacity. PubDate: Wed, 26 Sep 2018 09:26:18 +000
Abstract: This paper applies deep convolutional neural network (CNN) to identify tomato leaf disease by transfer learning. AlexNet, GoogLeNet, and ResNet were used as backbone of the CNN. The best combined model was utilized to change the structure, aiming at exploring the performance of full training and fine-tuning of CNN. The highest accuracy of 97.28% for identifying tomato leaf disease is achieved by the optimal model ResNet with stochastic gradient descent (SGD), the number of batch size of 16, the number of iterations of 4992, and the training layers from the 37 layer to the fully connected layer (denote as “fc”). The experimental results show that the proposed technique is effective in identifying tomato leaf disease and could be generalized to identify other plant diseases. PubDate: Wed, 26 Sep 2018 00:00:00 +000
Abstract: There are serious distortion problems in the history audio and video data. In view of the characteristics of audio data repair, the intelligent technology of audio evaluation is explored. As the traditional audio subjective evaluation method requires a large number of personal to audition and evaluation, the tester’s subjective sense of hearing deviation and sample space data limited the impact of the accuracy of the experiment. Based on the deep learning network, this paper designs an objective quality evaluation system for historical audio and video data and evaluates the performance of the system and the audio signal quality from the perspective of feature extraction and network parameter selection. Experiments show that the system has good performance in this experiment; the predictive results and subjective evaluation of the correlation and dispersion indicators are good, up to 0.91 and 0.19. PubDate: Tue, 18 Sep 2018 00:00:00 +000
Abstract: Active contour methods are widely used for medical image segmentation. Using level set algorithms the applications of active contour methods have become flexible and convenient. This paper describes the evaluation of the performance of the active contour models using performance metrics and statistical analysis. We have implemented five different methods for segmenting the synovial region in arthritis affected ultrasound image. A comparative analysis between the methods of segmentation was performed and the best segmentation method was identified using similarity criteria, standard error, and F-test. For further analysis, classification of the segmentation techniques using support vector machine (SVM) classifier is performed to determine the absolute method for synovial region detection. With these results, localized region based active contour named Lankton method is defined to be the best segmentation method. PubDate: Sun, 16 Sep 2018 00:00:00 +000
Abstract: Image segmentation is an essential task in computer vision and pattern recognition. There are two key challenges for image segmentation. One is to find the most discriminative image feature set to get high-quality segments. The other is to achieve good performance among various images. In this paper, we firstly propose a selective feature fusion algorithm to choose the best feature set by evaluating the results of presegmentation. Specifically, the proposed method fuses selected features and applies the fused features to region growing segmentation algorithm. To get better segments on different images, we further develop an algorithm to change threshold adaptively for each image by measuring the size of the region. The adaptive threshold can achieve better performance on each image than fixed threshold. Experimental results demonstrate that our method improves the performance of traditional region growing by selective feature fusion and adaptive threshold. Moreover, our proposed algorithm obtains promising results and outperforms some popular approaches. PubDate: Sun, 09 Sep 2018 00:00:00 +000
Abstract: The purposes of the algorithm presented in this paper are to select features with the highest average separability by using the random forest method to distinguish categories that are easy to distinguish and to select the most divisible features from the most difficult categories using the weighted entropy algorithm. The framework is composed of five parts: random samples selection with probabilistic output initial random forest classification processing based on the number of votes; semisupervised classification, which is an improvement of the supervision classification of random forest based on the weighted entropy algorithm; precision evaluation; and a comparison with the traditional minimum distance classification and the support vector machine (SVM) classification. In order to verify the universality of the proposed algorithm, two different data sources are tested, which are AVIRIS and Hyperion data. The results show that the overall classification accuracy of AVIRIS data is up to 87.36%, the kappa coefficient is up to 0.8591, and the classification time is 22.72s. Hyperion data is up to 99.17%, the kappa coefficient is up to 0.9904, and the classification time is 8.16s. Classification accuracy is obviously improved and efficiency is greatly improved, compared with the minimum distance and the SVM classifier and the CART classifier. PubDate: Tue, 04 Sep 2018 08:53:42 +000
Abstract: We propose an anomaly detection approach by learning a generative model using deep neural network. A weighted convolutional autoencoder- (AE-) long short-term memory (LSTM) network is proposed to reconstruct raw data and perform anomaly detection based on reconstruction errors to resolve the existing challenges of anomaly detection in complicated definitions and background influence. Convolutional AEs and LSTMs are used to encode spatial and temporal variations of input frames, respectively. A weighted Euclidean loss is proposed to enable the network to concentrate on moving foregrounds, thus restraining background influence. Moving foregrounds are segmented from the input frames using robust principal component analysis decomposition. Comparisons with state-of-the-art approaches indicate the superiority of our approach in anomaly detection. Generalization of anomaly detection is improved by enforcing the network to focus on moving foregrounds. PubDate: Mon, 03 Sep 2018 09:03:14 +000
Abstract: It is a challenging issue to deal with kinds of appearance variations in visual tracking. Existing tracking algorithms build appearance models upon target templates. Those models are not robust to significant appearance variations due to factors such as illumination variations, partial occlusions, and scale variation. In this paper, we propose a robust tracking algorithm with a learnt dictionary to represent target candidates. With the learnt dictionary, a target candidate is represented with a linear combination of dictionary atoms. The discriminative information in learning samples is exploited. In the meantime, the learning processing of dictionaries can learn appearance variations. Based on the learnt dictionary, we can get a more stable representation for target candidates. Additionally, the observation likelihood is evaluated based on both the reconstruct error and dictionary coefficients with constraint. Comprehensive experiments demonstrate the superiority of the proposed tracking algorithm to some state-of-the-art tracking algorithms. PubDate: Sun, 02 Sep 2018 00:00:00 +000
Abstract: Automatic audio announcement systems are widely used in public places such as transportation vehicles and facilities, hospitals, and banks. However, these systems cannot be used by people with hearing impairment. That brings great inconvenience to their lives. In this paper, an approach of audio announcement detection and recognition for the hearing-impaired people based on the smart phone is proposed and a mobile phone application (app) is developed, taking the bank as a major applying scenario. Using the app, the users can sign up alerts for their numbers and then the system begins to detect audio announcements using the microphone on the smart phone. For each audio announcement detected, the speech within it is recognized and the text is displayed on the screen of the phone. When the number the user input is announced, alert will be given by vibration. For audio announcement detection, a method based on audio segment classification and postprocessing is proposed, which uses a SVM classifier trained on audio announcements and environment noise collected in banks. For announcement speech recognition, an ASR engine is developed using a GMM-HMM-based acoustic model and a finite state transducer (FST) based grammar. The acoustic model is trained on audio announcement speech collected in banks, and the grammar is human-defined according to the patterns used by the automatic audio announcement systems. Experimental results show that character error rates (CERs) around 5% can be achieved for the announcement speech, which shows feasibility of the proposed method and system. PubDate: Thu, 16 Aug 2018 06:53:59 +000
Abstract: Lane detection is a challenging problem. It has attracted the attention of the computer vision community for several decades. Essentially, lane detection is a multifeature detection problem that has become a real challenge for computer vision and machine learning techniques. Although many machine learning methods are used for lane detection, they are mainly used for classification rather than feature design. But modern machine learning methods can be used to identify the features that are rich in recognition and have achieved success in feature detection tests. However, these methods have not been fully implemented in the efficiency and accuracy of lane detection. In this paper, we propose a new method to solve it. We introduce a new method of preprocessing and ROI selection. The main goal is to use the HSV colour transformation to extract the white features and add preliminary edge feature detection in the preprocessing stage and then select ROI on the basis of the proposed preprocessing. This new preprocessing method is used to detect the lane. By using the standard KITTI road database to evaluate the proposed method, the results obtained are superior to the existing preprocessing and ROI selection techniques. PubDate: Tue, 07 Aug 2018 00:00:00 +000
Abstract: Scene understanding is to predict a class label at each pixel of an image. In this study, we propose a semantic segmentation framework based on classic generative adversarial nets (GAN) to train a fully convolutional semantic segmentation model along with an adversarial network. To improve the consistency of the segmented image, the high-order potentials, instead of unary or pairwise potentials, are adopted. We realize the high-order potentials by substituting adversarial network for CRF model, which can continuously improve the consistency and details of the segmented semantic image until it cannot discriminate the segmented result from the ground truth. A number of experiments are conducted on PASCAL VOC 2012 and Cityscapes datasets, and the quantitative and qualitative assessments have shown the effectiveness of our proposed approach. PubDate: Sun, 05 Aug 2018 07:06:01 +000
Abstract: Visual tracking is a challenging research topic in the field of computer vision with many potential applications. A large number of tracking methods have been proposed and achieved designed tracking performance. However, the current state-of-the-art tracking methods still can not meet the requirements of real-world applications. One of the main challenges is to design a good appearance model to describe the target’s appearance. In this paper, we propose a novel visual tracking method, which uses compressed features to model target’s appearances and then uses SVM to distinguish the target from its background. The compressed features were obtained by the zero-tree coding on multiscale wavelet coefficients extracted from an image, which have both the low dimensionality and discriminate ability and therefore ensure to achieve better tracking results. The experimental comparisons with several state-of-the-art methods demonstrate the superiority of the proposed method. PubDate: Wed, 01 Aug 2018 00:00:00 +000
Abstract: With the rapidly growing number of images over the Internet, efficient scalable semantic image retrieval becomes increasingly important. This paper presents a novel approach for semantic image retrieval by combining Convolutional Neural Network (CNN) and Markov Random Field (MRF). As a key step, image concept detection, that is, automatically recognizing multiple semantic concepts in an unlabeled image, plays an important role in semantic image retrieval. Unlike previous work that uses single-concept classifiers one by one, we detect semantic multiconcept by using a multiconcept scene classifier. In other words, our approach takes multiple concepts as a holistic scene for multiconcept scene learning. Specifically, we first train a CNN as a concept classifier, which further includes two types of classifiers: a single-concept fully connected classifier that is best suited to single-concept detection and a multiconcept scene fully connected classifier that is good for holistic scene detection. Then we propose an MRF-based late fusion approach that is able to effectively learn the semantic correlation between the single-concept classifier and multiconcept scene classifier. Finally, the semantic correlation among the subconcepts of images is cought to further improve detection precision. In order to investigate the feasibility and effectiveness of our proposed approach, we conduct comprehensive experiments on two publicly available image databases. The results show that our proposed approach outperforms several state-of-the-art approaches. PubDate: Wed, 01 Aug 2018 00:00:00 +000
Abstract: This paper addresses the problem of adaptive power control based on outage probability minimization in Vehicular Ad Hoc Networks (VANETs), called a Power Control Algorithm Based on Outage Probability Awareness (PC-OPA). Unlike most of the existing works, our power control method aims at minimizing the outage probability and then is subject to the density of nodes in certain area. To fulfill power control, cumulative interference is assumed to be available at the transmitter of each terminal. The transmitters sent data by maximum power and then get the cumulative interference-aware outage probability. Furthermore, we build the interference model by stochastic geometric theory and then derive the expression between outage probability and cumulative interference. According to the expression, we adjust the transmitter power and optimize the outage probability. Simulation results are provided to demonstrate the effectiveness of the proposed power control strategies. It is shown that the PC-OPA can achieve a significant performance gain in terms of the outage probability and throughputs. Comparing MPC (Maximum Power Control algorithm) and WFPC (Water-Filled Power Control algorithm), the proposed PC-OPA decreased by 23% in terms of the outage probability and increased by 25% in terms of throughputs. PubDate: Wed, 01 Aug 2018 00:00:00 +000
Abstract: Speech synthesis is an important research content in the field of human-computer interaction and has a wide range of applications. As one of its branches, singing synthesis plays an important role. Beijing Opera is a famous traditional Chinese opera, and it is called Chinese quintessence. The singing of Beijing Opera carries some features of speech but it has its own unique pronunciation rules and rhythms which differ from ordinary speech and singing. In this paper, we propose three models for the synthesis of Beijing Opera. Firstly, the speech signals of the source speaker and the target speaker are extracted by using the straight algorithm. And then through the training of GMM, we complete the voice control model to input the voice to be converted and output the voice after the voice conversion. Finally, by modeling the fundamental frequency, duration, and frequency separately, a melodic control model is constructed using GAN to realize the synthesis of the Beijing Opera fragment. We connect the fragments and superimpose the background music to achieve the synthesis of Beijing Opera. The experimental results show that the synthesized Beijing Opera has some audibility and can basically complete the composition of Beijing Opera. We also extend our models to human-AI cooperative music generation: given a target voice of human, we can generate a Beijing Opera which is sung by a new target voice. PubDate: Tue, 17 Jul 2018 00:00:00 +000
Abstract: Person re-identification, aiming to identify the same pedestrian images across disjoint camera views, is a key technique of intelligent video surveillance. Although existing methods have developed both theories and experimental results, most of effective ones pertain to fully supervised training styles, which suffer the small sample size (SSS) problem a lot, especially in label-insufficient practical applications. To bridge SSS problem and learning model with small labels, a novel semisupervised co-metric learning framework is proposed to learn a discriminative Mahalanobis-like distance matrix for label-insufficient person re-identification. Different from typical co-training task that contains multiview data originally, single-view person images are firstly decomposed into pseudo two views, and then metric learning models are produced and jointly updated based on both pseudo-labels and references iteratively. Experiments carried out on three representative person re-identification datasets show that the proposed method performs better than state of the art and possesses low label sensitivity. PubDate: Sun, 15 Jul 2018 00:00:00 +000
Open Access journal
[338 journals]
