Abstract Since residual learning was proposed, identity mapping has been widely utilized in various neural networks. The method enables information transfer without any attenuation, which plays a ...significant role in training deeper networks. However, interference with unhindered transmission also affects the network’s performance. Accordingly, we propose a generalized residual learning architecture called reverse attention (RA), which applies high-level semantic features to supervise low-level information in the identity mapping branch. It means that higher semantic features selectively transmit low-level information to deeper layers. In addition, we propose a Modified Global Response Normalization(M-GRN) to implement reverse attention. RA-Net is derived by embedding M-GRN in the residual learning framework. The experiments show that the RA-Net brings significant improvements over residual networks on typical computer vision tasks. For classification on ImageNet-1K, compared with resnet101, RA-Net improves the Top-1 accuracy by 1.7% with comparable parameters and computational cost. For COCO detection, on Faster R-CNN, reverse attention improves box AP by 1.9%. Meanwhile, reverse attention improves UpperNet’s mIoU by 0.7% on ADE20K segmentation.
Classic detectors divide the candidate boxes into positive and negative groups based on their intersection-over-union (IoU) with matched objects. Such a sharp label assignment method does not ...directly consider the distance between the centers of the two boxes. Consequently, some positive samples are ambiguous, which limits the detection performance. In this paper, we propose a new label assignment method by considering two different perspectives. The first perspective is IoU to indicate the degree of overlap. The other one is the defined variable to characterize the center-distance between the candidate box and its matched ground truth. In addition, the classification is usually optimized by Focal Loss for paying more attention to hard examples, but it affects the training of high-quality samples. Therefore, we define the Soft Focal Loss (SFL) and the quality factor that reflects the quality of the samples. Embedding the quality factor into SFL makes the network focus on learning high-quality rather than hard examples. Furthermore, the quality factor is utilized to re-weight the classification and regression losses to enhance the correlation between these two tasks. Experiments on COCO show that the proposed approach can improve RetinaNet by 1.3% and 1.2% AP with backbone ResNet-50 and ResNet-101 in 1x training schedule, without incurring any additional overhead.
In this article, a real-time vehicle sideslip angle state observer is proposed, which is based on the EKF algorithm. Firstly, based on a 2-DOF dynamical model and the tire lateral force model, the ...vehicle sideslip angle state observer model with a self-adapted truncation procedure is established by combining the EKF and the least squares methods. The calculation of the Jacobi matrix in the time domain is transformed into a calculation in the frequency domain. A self-adapted update noise estimation method and an initial value setting strategy are proposed as well. Finally, a hardware-in-the-loop simulation is carried out by Matlab/Simulink-CarSim-dSPACE, and the real-time reliability of the estimation method is verified and analyzed by RMSE.
In certain cases, the condition of the fetus can be revealed by the fetal heart sound. However, when the sound is detected, it is mixed with noise from the external environment as well as internal ...disturbances. Our exclusive sensor, which was constructed of copper with an enclosed cavity, was designed to prevent external noise. In the sensor, a polyvinylidene fluoride (PVDF) piezoelectric film, with a frequency range covering that of the fetal heart sound, was adopted to convert the sound into an electrical signal. The adaptive support vector regression (SVR) algorithm was proposed to reduce internal disturbance. The weighted-index average algorithm with deviation correction was proposed to calculate the fetal heart rate. The fetal heart sound data were weighted automatically in the window and the weight was modified with an exponent between windows. The experiments show that the adaptive SVR algorithm was superior to empirical mode decomposition (EMD), the self-adaptive least square method (LSM), and wavelet transform. The weighted-index average algorithm weakens fetal heart rate jumps and the results are consistent with reality.
Estimation of the age of human bloodstains is of great importance in forensic practices, but it is a challenging task because of the lack of a well-accepted, reliable, and established method. Here, ...the attenuated total reflection (ATR)-Fourier transform infrared (FTIR) technique combined with advanced chemometric methods was utilized to determine the age of indoor and outdoor bloodstains up to 107 days. The bloodstain storage conditions mimicked crime scene scenarios as closely as possible. Two partial least squares regression models-indoor and outdoor models with 7-85 days-exhibited good performance for external validation, with low values of predictive root mean squared error (5.83 and 4.77) and high R
values (0.94 and 0.96) and residual predictive deviation (4.08 and 5.14), respectively. Two partial least squares-discriminant analysis classification models were built and demonstrated excellent distinction between fresh (age ≤1 d) and older (age >1 d) bloodstains, which is highly valuable for forensic investigations. These findings demonstrate that ATR-FTIR spectroscopy coupled with advanced chemometric methods can be employed as a rapid and non-destructive tool for age estimation of bloodstains in real-world forensic investigation.
Traffic accidents due to fatigue account for a large proportion of road fatalities. Based on simulated driving experiments with drivers recruited from college students, this paper investigates the ...use of heart rate variability (HRV) features to detect driver fatigue while considering sex differences. Sex-independent and sex-specific differences in HRV features between alert and fatigued states derived from 2 min electrocardiogram (ECG) signals were determined. Then, decision trees were used for driver fatigue detection using the HRV features of either all subjects or those of only males or females. Nineteen, eighteen, and thirteen HRV features were significantly different (Mann–Whitney U test, p < 0.01) between the two mental states for all subjects, males, and females, respectively. The fatigue detection models for all subjects, males, and females achieved classification accuracies of 86.3%, 94.8%, and 92.0%, respectively. In conclusion, sex differences in HRV features between drivers’ mental states were found according to both the statistical analysis and classification results. By considering sex differences, precise HRV feature-based driver fatigue detection systems can be developed. Moreover, in contrast to conventional methods using HRV features from 5 min ECG signals, our method uses HRV features from 2 min ECG signals, thus enabling more rapid driver fatigue detection.
Estimating PMI is of great importance in forensic investigations. Although many methods are used to estimate the PMI, a few investigations focus on the postmortem redistribution. In this study, ...ultraviolet-visible (UV-Vis) measurement combined with visual inspection indicated a regular diffusion of hemoglobin into plasma after death showing the redistribution of postmortem components in blood. Thereafter, attenuated total reflection-Fourier transform infrared (ATR-FTIR) spectroscopy was used to confirm the variations caused by this phenomenon. First, full-spectrum partial least-squares (PLS) and genetic algorithm combined with PLS (GA-PLS) models were constructed to predict the PMI. The performance of GA-PLS model was better than that of full-spectrum PLS model based on its root mean square error (RMSE) of cross-validation of 3.46 h (R2 = 0.95) and the RMSE of prediction of 3.46 h (R2 = 0.94). The investigation on the similarity of spectra between blood plasma and formed elements also supported the role of redistribution of components in spectral changes in postmortem plasma. These results demonstrated that ATR-FTIR spectroscopy coupled with the advanced mathematical methods could serve as a convenient and reliable tool to study the redistribution of postmortem components and estimate the PMI.
Regarding the set of all feature attributes in a given database as the universal set, this monograph discusses various nonadditive set functions that describe the interaction among the contributions ...from feature attributes towards a considered target attribute. Then, the relevant nonlinear integrals are investigated. These integrals can be applied as aggregation tools in information fusion and data mining, such as synthetic evaluation, nonlinear multiregressions, and nonlinear classifications. Some methods of fuzzification are also introduced for nonlinear integrals such that fuzzy data can be treated and fuzzy information is retrievable.The book is suitable as a text for graduate courses in mathematics, computer science, and information science. It is also useful to researchers in the relevant area.Sample Chapter(s)Chapter 1: Introduction (114 KB)Contents: Basic Knowledge on Classical SetsFuzzy SetsSet FunctionsIntegrationsInformation FusionOptimization and Soft ComputingIdentification of Set FunctionsMultiregressions Based on Nonlinear IntegralsClassifications Based on Nonlinear IntegralsData Mining with Fuzzy DataReadership: Graduate students and research students interested in mathematics and computer science.
Display omitted
•FTIR spectroscopy was utilized to investigate the postmortem changes of human skeletal remains.•The environmental effects to postmortem changes were considered and ...studied.•Chemometric models help to estimate the late postmortem interval.
Due to a lack of reliable and accurate methods, determining the postmortem interval (PMI) of human skeletal remains is one of the most important and challenging tasks in forensic medicine. In this paper, we studied the changes to bone chemistry with increasing PMI in two different experimental conditions using Fourier transform infrared (FTIR) spectroscopy in conjunction with chemometrics methods Paired bone samples collected from 56 human corpses were buried (placed in soil) and unburied (exposed to the air) for intervals between 76 and 552 days. The results of principle component analysis (PCA) showed the chemical differences of these two cases had a significant influence on the rate of decomposition of the remains. Meanwhile, satisfactory predictions were performed by the genetic algorithm combined with partial least-squares (GA-PLS) with the root mean square errors of prediction (RMSEP) of 50.93days for buried bones and 71.03days for unburied bones. Moreover, the amide I region of proteins and the area around 1390cm−1, which is associated with fatty acids, were identified with regular changes by GA-PLS and played an important role in estimating PMI. This study illustrates the feasibility of utilizing FTIR spectroscopy and chemometrics as an attractive alternative for estimating PMI of human remains and the great potential of these techniques in real forensic cases with natural conditions.