Search

Automatic detection of epileptic seizures is important in epilepsy control and treatment, and specific feature extraction assists in accurate detection. We developed a feature extraction method for seizure detection based on multi-site synchronous changes and an edge detection algorithm. We investigated five chronic temporal lobe epilepsy rats with 8- and 12-channel detection sites in the hippocampus and limbic system. Multi-site synchronous changes were selected as a specific feature and implemented as a seizure detection method. For preprocessing, we used magnitude-squared coherence maps and Canny edge detection algorithm to find the frequency band with the most significant change in synchronization and the important channel pairs. In detection, we used the maximal cross-correlation coefficient as an indicator of synchronization and the correlation coefficient curves' average value and standard deviation as two detection features. The method achieved high performance, with an average 96.60% detection rate, 2.63/h false alarm rate, and 1.25 s detection delay. The experimental results show that synchronization is an appropriate feature for seizure detection. The magnitude-squared coherence map can assist in selecting a specific frequency band and channel pairs to enhance the detection result. We found that individuals have a specific frequency band that reflects the most significant synchronization changes, and our method can individually adjust parameters and has good detection performance. [ABSTRACT FROM AUTHOR]

Since the human face preserves the richest information for recognizing individuals, face recognition has been widely investigated and achieved great success in various applications in the past decades. However, face spoofing attacks (e.g., face video replay attack) remain a threat to modern face recognition systems. Though many effective methods have been proposed for anti-spoofing, we find that the performance of many existing methods is degraded by illuminations. It motivates us to develop illumination-invariant methods for anti-spoofing. In this paper, we propose a two-stream convolutional neural network (TSCNN), which works on two complementary spaces: RGB space (original imaging space) and multi-scale retinex (MSR) space (illumination-invariant space). Specifically, the RGB space contains the detailed facial textures, yet it is sensitive to illumination; MSR is invariant to illumination, yet it contains less detailed facial information. In addition, the MSR images can effectively capture the high-frequency information, which is discriminative for face spoofing detection. Images from two spaces are fed to the TSCNN to learn the discriminative features for anti-spoofing. To effectively fuse the features from two sources (RGB and MSR), we propose an attention-based fusion method, which can effectively capture the complementarity of two features. We evaluate the proposed framework on various databases, i.e., CASIA-FASD, REPLAY-ATTACK, and OULU, and achieve very competitive performance. To further verify the generalization capacity of the proposed strategies, we conduct cross-database experiments, and the results show the great effectiveness of our method. [ABSTRACT FROM AUTHOR]

In this paper, we present novel sharp attention networks by adaptively sampling feature maps from convolutional neural networks for person re-identification (re-ID) problems. Due to the introduction of sampling-based attention models, the proposed approach can adaptively generate sharper attention-aware feature masks. This greatly differs from the gating-based attention mechanism that relies on soft gating functions to select the relevant features for person re-ID. In contrast, the proposed sampling-based attention mechanism allows us to effectively trim irrelevant features by enforcing the resultant feature masks to focus on the most discriminative features. It can produce sharper attentions that is more assertive in localizing subtle features relevant to re-identifying people across cameras. For this purpose, a differentiable Gumbel-Softmax sampler is employed to approximate the Bernoulli sampling to train the sharp attention networks. Extensive experimental evaluations demonstrate the superiority of this new sharp attention model for person re-ID over other related existing, published state-of-the-art works on three challenging benchmarks, including CUHK03, Market-1501, and DukeMTMC-reID. [ABSTRACT FROM AUTHOR]

We propose a deep convolutional neural network (CNN) for general no-reference image quality assessment (NR-IQA), i.e., accurate prediction of image quality without a reference image. The proposed model consists of three components such as a local feature extractor that is a fully CNN, an encoding module with an inherent dictionary that aggregates local features to output a fixed-length global quality-aware image representation, and a regression module that maps the representation to an image quality score. Our model can be trained in an end-to-end manner, and all of the parameters, including the weights of the convolutional layers, the dictionary, and the regression weights, are simultaneously learned from the loss function. In addition, the model can predict quality scores for input images of arbitrary sizes in a single step. We tested our method on commonly used image quality databases and showed that its performance is comparable with that of state-of-the-art general-purpose NR-IQA algorithms. © 2018 SPIE and IS&T [DOI: 10.1117/1.JEI.27.2.023020] [ABSTRACT FROM AUTHOR]

• Magnitude-squared coherence method was used to find the frequency band and channels. • Hypergraph feature coefficients were used to distinguish different signals. • Machine learning methods were used to classify the seizure signals. • Individually adjust parameters for different patients adaptively. Effective feature extraction is the key to successful seizure detection. A good feature should be adaptable, noise-resistant, and informative. In this study, a seizure detection method based on hypergraph features and machine learning was proposed. During the preprocessing step, the magnitude-squared coherence (MSC) method was used to select the useful channels and frequency band. The hypergraph method was introduced to represent the correlation network based on the MSC result. Hypergraphs can show relationships between channels, and they have higher dimensional properties than normal graphs. To quantify the characteristics of these channels, three feature coefficients, which are HG-Independent(HG-Ind), HG-Dependent(HG-Dep), and HG-Share(HG-Sha), were derived and input as features into the machine learning algorithm. Two machine learning algorithms, random forest (RF) and support vector machine (SVM), were tested on the data of eight patients in our study. According to the results, RF had a better performance than SVM, and the group of three features performed better than the single feature. Three additional methods from other papers, extra-trees, random forest based on spectral correlation, and neural networks, were compared with the HG-RF, and the HG-RF had the best performance, with an accuracy of 96.30 %. This method showed resistance to noise and adaptability for different patient data. In addition, this method facilitated a balance between effectiveness and efficiency and provided a new idea for online detection of epilepsy. [ABSTRACT FROM AUTHOR]

The new video coding standard, High Efficiency Video Coding (HEVC), achieves much higher coding efficiency than the state-of-the-art H.264. Transcoding H.264 video to HEVC video is important to enable gradual migration to HEVC. Therefore, a fast H.264 to HEVC transcoding algorithm based on region feature analysis is proposed. First, each frame is segmented into three regions in units of coding tree unit (CTU) based on the correlation between image coding complexities and coding bits of the H.264 source stream. Then the searching depth range of each CTU is adaptively decided according to the region type. After that, motion vectors are de-noise filtered and clustered in order to analyze the region features of coding unit (CU). Based on the analysis results, the minimum searching depth of CU and partitions of prediction unit (PU) are optimally selected, and the motion vector predictor and search window size of motion estimation are also optimally decided for further reduction of the computational complexity. Experimental results show that the proposed algorithm achieves a significant improvement on transcoding speed, while maintaining high Rate-Distortion performance. [ABSTRACT FROM AUTHOR]

Deep convolutional neural networks (CNNs) have achieved considerable success with image denoising. However, they still lack consistent performance across different noise types and levels. We extend noise scenarios to four categories: Gaussian, random-impulse, salt-and-pepper, and Poisson. We also propose a multinoise-type blind denoising network (MBDNet) that solves the blind denoising task using a uniform deep CNN architecture. The network can be divided into two stages where a concise CNN is first used to estimate auxiliary noise-type and noise-level information. Estimation results are then integrated as additional channels of the noisy image and are fed to the subsequent denoising stage. A unique two-branch structure is further adopted in the residual denoising CNN, wherein a shallow branch predicts the filter-flow mask and adaptively adjusts the feature extraction of the parallel deep branch. Extensive experiments on synthetic noisy images validate the effectiveness of the noise-estimation and denoising subnetworks and show that MBDNet is highly competitive as compared to state-of-the-art methods in both denoising performance and model runtime. [ABSTRACT FROM AUTHOR]