Infrared thermal imaging technology has been adopted for its advantages such as fast detection response, noncontact testing, and applicability to various objects. When testing a target object, this technology presents...
详细信息
Infrared thermal imaging technology has been adopted for its advantages such as fast detection response, noncontact testing, and applicability to various objects. When testing a target object, this technology presents temperature distribution information of the target object's surface in the form of an image, achieving visualization. However, due to limitations in the hardware system of the thermal imager and the noise generated during the detection process, the resolution of infrared images is relatively low, and the details of the image are not rich enough, leading to limitations in specific defect detection. In this study, a defect super-resolution algorithm based on infrared thermal imaging physical kernel is proposed. The imaging degradation factors of the infrared images are analyzed, and the modulation transfer function of the infrared thermal imaging system is used as the physical prior to generate the underlying blur kernel of the infrared images. The infrared images are then reconstructed using a super-resolution algorithm based on the blur kernel. Obtained experimental results have demonstrated that the proposed method significantly improves the defect detection rate and the overall image quality. The demo code will be updated soon in https://***/gaobin/zh_CN/lwcg/153392/list/***.
With the integrated sensing and communications (ISAC) proposed and the development of the internet of things (IoT), indoor localization and sensing have become a hot spot of current research. Most of these researches ...
详细信息
A method to construct a large transmission matrix (TM) of highly scattering media based on an algorithm is presented. For imaging through scattering media, measuring a large TM enables image reconstruction with high s...
详细信息
A method to construct a large transmission matrix (TM) of highly scattering media based on an algorithm is presented. For imaging through scattering media, measuring a large TM enables image reconstruction with high spatial resolution;however, it is difficult to operate in practice because the time required for the calibration of the TM is proportional to the number of calibrated input modes. Instead of directly achieving calibration using advanced hardware, a large TM is defined and constructed based on a multiframe image super-resolution algorithm, which traditionally aims to recover a high-resolution (HR) image of the original object from several low-resolution images. With sufficient size improvement, the defined large TM is constructed from several naturally measured complementary small TMs, each of them is measured using a subpixel-shifted projector phase mask. The realized large TM, which is constructed through subpixel registration and interpolation operations at the help of a hypothetical HR mask, enables HR image reconstruction from the original object's speckle signal. The feasibility of the proposed method is proven via optical experiments involving statistical analysis and image reconstruction. The proposed method benefits existing TM-based image reconstruction applications and offers a new perspective on the size of TM measurements and the imaging resolution.
Researchers who have successfully applied convolutional neural networks in the field of computer vision have developed a series of image super-resolution algorithms based on convolutional neural networks. In this way,...
详细信息
Researchers who have successfully applied convolutional neural networks in the field of computer vision have developed a series of image super-resolution algorithms based on convolutional neural networks. In this way, the effect of image reconstruction is greatly improved. The existing image super-resolution algorithm realizes the ideal mathematical modeling of the image degradation process, and obtains a low-resolution image through a properly modeled degradation process. Then, a pair of high-resolution images and low-resolution images is used for surveillance-related training. As an effective software processing technology, image super-resolution reconstruction can break the physical boundaries of image equipment and improve the spatial resolution of the acquired image. This thesis focuses on the research of image emphasis and super-resolution reconstruction based on space animation, especially the image processing of the position and docking of spacecraft. Aiming at the characteristics of spatial moving images and the problems of application programs such as edge patterns, noise effects, and real-time processing, the algorithm is studied. We propose an image enhancement and super-resolution reconstruction algorithm based on multi-scale conversion. This research and design method once again realizes the image design and reconstruction.
Much research has been conducted to improve the defect-detection rate and detection accuracy of the imaging technology used in terahertz nondestructive testing. Due to the power limit of light sources and noise interf...
详细信息
Much research has been conducted to improve the defect-detection rate and detection accuracy of the imaging technology used in terahertz nondestructive testing. Due to the power limit of light sources and noise interference in terahertz equipment, images have low resolution and fuzzy defect edges. Hence, improving the resolution is crucial for detecting defects. We designed an edge detection network structure based on a traditional deep neural network. Besides, we devised a node-fusing strategy to train the network. It demonstrates significant improvement of the resolution of the terahertz defect contour. A quartz fiber composites with embedded defects was tested with our network. The results showed that the proposed super-resolution reconstruction algorithm improves resolution, particularly on the edges of defect contours.
In recent years, Wi-Fi has become the main gateway that connects users to the Internet. Considering the availability of Wi-Fi signals, and their suitability for channel estimation, IEEE established the Wi-Fi Sensing (...
详细信息
In recent years, Wi-Fi has become the main gateway that connects users to the Internet. Considering the availability of Wi-Fi signals, and their suitability for channel estimation, IEEE established the Wi-Fi Sensing (WS) Task Group whose purpose is to study the feasibility of Wi-Fi-based environment sensing. However, Wi-Fi signals are transmitted over limited bandwidths with a relatively small number of antennas in bursts, fundamentally limiting the range, Angle-of-Arrival and speed resolutions. This paper presents a super-resolution algorithm to perform the parameter estimation in a quasi-monostatic WS scenario. The proposed algorithm, RIVES, estimates the range, Angle-of-Arrival and speed parameters with Vandermonde decomposition of Hankel matrices. To estimate the size of the signal subspace, RIVES uses a novel Model Order Selection method which eliminates spurious noise targets based on their distance to the noise and signal subspaces. Various scenarios with multiple targets are simulated to show the robustness of RIVES. In order to prove its accuracy, real-life indoor experiments are conducted with human targets by using Software Defined Radios.
Fetal Magnetic Resonance Imaging (MRI) is an important noninvasive diagnostic tool to characterize the central nervous system (CNS) development, significantly contributing to pregnancy management. In clinical practice...
详细信息
Fetal Magnetic Resonance Imaging (MRI) is an important noninvasive diagnostic tool to characterize the central nervous system (CNS) development, significantly contributing to pregnancy management. In clinical practice, fetal MRI of the brain includes the acquisition of fast anatomical sequences over different planes on which several biometric measurements are manually extracted. Recently, modern toolkits use the acquired two-dimensional (2D) images to reconstruct a super-resolution (SR) isotropic volume of the brain, enabling three-dimensional (3D) analysis of the fetal *** analyzed 17 fetal MR exams performed in the second trimester, including orthogonal T2-weighted (T2w) Turbo Spin Echo (TSE) and balanced Fast Field Echo (b-FFE) sequences. For each subject and type of sequence, three distinct high-resolution volumes were reconstructed via NiftyMIC, MIALSRTK, and SVRTK toolkits. Fifteen biometric measurements were assessed both on the acquired 2D images and SR reconstructed volumes, and compared using Passing-Bablok regression, Bland-Altman plot analysis, and statistical *** indicate that NiftyMIC and MIALSRTK provide reliable SR reconstructed volumes, suitable for biometric assessments. NiftyMIC also improves the operator intraclass correlation coefficient on the quantitative biometric measures with respect to the acquired 2D images. In addition, TSE sequences lead to more robust fetal brain reconstructions against intensity artifacts compared to b-FFE sequences, despite the latter exhibiting more defined anatomical *** findings strengthen the adoption of automatic toolkits for fetal brain reconstructions to perform biometry evaluations of fetal brain development over common clinical MR at an early pregnancy stage.
Integrated sensing and communication (ISAC) systems have been investigated by the research and standardization communities in the recent past. Accurately localizing the target and tracking the target's movement ar...
详细信息
ISBN:
(纸本)9798350348439;9798350384611
Integrated sensing and communication (ISAC) systems have been investigated by the research and standardization communities in the recent past. Accurately localizing the target and tracking the target's movement are critical for numerous smart Internet of Things (IoT) systems (smart manufacturing, smart transportation, etc.). This paper aims to realize super-resolution localization and tracking in WiFi sensing by leveraging the IEEE 802.11ad beamforming training procedure. We leverage the CLEAN-Space-Alternating Generalized Expectation-maximization (CLEAN-SAGE) algorithm on a single beam sweeping cycle for target localization and investigate the targets' delays and angle estimation. For tracking moving targets, we design mechanisms to estimate the target's motion, including the target's velocity and motion pattern, such as estimating the target's spatial positions over time to obtain the Doppler shift or tracking its trajectory using a Kalman filter. In order to prove that our approach works effectively, we conduct a thorough performance evaluation study. Our evaluation results confirm that the CLEAN-SAGE algorithm can achieve estimation performance beyond the ISAC system's inherent bandwidth and beamwidth constraints. Furthermore, we provide insights into how system configurations, including antenna size, beam overlap, and the number of iterations in the SAGE algorithm, influence its performance.
During oil spills, using ultrasound to measure the thickness of thick oil slick can provide critical information for determining the oil spill response strategy and improving the response efficiency. However, the ultr...
详细信息
During oil spills, using ultrasound to measure the thickness of thick oil slick can provide critical information for determining the oil spill response strategy and improving the response efficiency. However, the ultrasonic measuring ability is strictly limited by the wavelength of ultrasound. To expand the ability to measure a thin oil slick thickness, a super-resolution algorithm based on a generalization of the sparse iterative covariance estimation (SPICE) algorithm was proposed to calculate the time of flight (TOF) by processing ultrasonic data collected with two transducers installed on a remotely operated vehicle (ROV). First, the cross correlation of ultrasonic signals was performed to improve the signal-to-noise ratio (SNR). Then, the outputs were processed with the inverse discrete Fourier transform (IDFT) to directly generate snapshot samples. At last, a sparse iterative covariance-based algorithm taking account of the signal sparseness was developed to deal with snapshot samples to achieve the TOF. The simulations were conducted to compare the proposed method with the generalized cross correlation (GCC) method and the modified orthogonal matching pursuit (MOMP) method. The experimental results show that the proposed method can significantly improve the ability to measure thin oil slick thickness by almost halving the minimum measurable thickness with superior measurement accuracy in comparison to the GCC method. It could be applied to measure the thickness of an oil slick during an oil spill response.
Ground localization systems based on cellular signals are vulnerable to the hazards of signal power attenuation and multipath propagation in urban environments. Non-coherent accumulation is an effective solution to th...
详细信息
Ground localization systems based on cellular signals are vulnerable to the hazards of signal power attenuation and multipath propagation in urban environments. Non-coherent accumulation is an effective solution to this problem, but its application to cellular localization systems has not been properly discussed. In this paper, we propose two cellular time-of-arrival (TOA) estimation methods based on non-coherent accumulation: the "TOA estimation algorithm based on non-coherent accumulation of the channel impulse response" (nch-CIR) in the time domain and the "superresolution TOA Estimation algorithm based on non-coherent accumulation of the covariance matrix" (nch-SRA) in the frequency domain. Among these two methods, the nch-CIR algorithm has a lower computational cost and better anti-noise performance, and the nch-SRA algorithm has better performance in terms of multipath delay estimation. Through theoretical analysis and extensive simulations, we also discuss the influence of mobility on these two methods. In addition, experiments are conducted to evaluate the performance of the proposed method using real collected cellular signals. The results show that both nch-CIR and nch-SRA can achieve a better performance compared with the conventional methods.
暂无评论