a kind of spatial-temporal neural network video smoke detection algorithm is proposed in order to solve the problems associated with the incorrect classification of the static approximate smoke background in the face ...
详细信息
ISBN:
(数字)9798350376548
ISBN:
(纸本)9798350376555
a kind of spatial-temporal neural network video smoke detection algorithm is proposed in order to solve the problems associated with the incorrect classification of the static approximate smoke background in the face of the detection of smoke in video detection networks, and the problem of false alarms and of the original test model algorithms being different in different detection environments. Based on the original YOLO v4 neural network algorithm, this paper introduces a k-means + + algorithm and genetic algorithm, while using the algorithm's clustering function to classify the sample points of the real boxes of the image data set, which make it a more suitable anchor. At the same time, the genetic algorithm is used to adjust its anchor in order to allow the generated anchor to adapt to the needs related to smoke detection. In the original neural network model, the dual-stream network model algorithm is used to extract information from the first step of the YOLO algorithm in order to further filter the smoke's characteristics as well as filter out error information, all to improve the detection capabilities of the overall neural network for video smoke fog images. Compared with traditional YOLOv4 networks, the algorithm obtained by the model algorithm has been improved by 8.51°/0. In actual tests, the alarm time requirements of the smoke alarm test program for early fire monitoring and the alarm systems for visual images were improved, and the detection accuracy of the network was also improved based on the assurance of the detection speed, while the performance of the model algorithm was also improved for different scenes.
Pipelines are known as a traditional solution for transporting various media such as gas, oil, and water. But pipes are always combined with the defects. Some of these defects are caused by the manufactory process, wh...
详细信息
The application of CBCT systems in intraoperative environments has become increasingly common, but concurrent CBCT systems are unsuitable for situations that require a large longitudinal imaging FoV, such as orthopedi...
详细信息
ISBN:
(数字)9798350371499
ISBN:
(纸本)9798350371505
The application of CBCT systems in intraoperative environments has become increasingly common, but concurrent CBCT systems are unsuitable for situations that require a large longitudinal imaging FoV, such as orthopedics. To increase longitudinal coverage, we developed a dual-source CBCT (DS-CBCT) system in which two ray sources are symmetrically placed along the central plane. After further analyzing its geometric characteristics, we propose an analytical reconstruction algorithm termed DT-FDK specialized for DS-CBCT, which combines cone-beam rebinning and two rays with smaller cone angles among the four conjugate rays to further suppress cone beam artifacts. The system design and reconstruction algorithm are tested on both simulated and real-scanned data. Results show that the DS-CBCT can expand the effective imaging volume by 37.1% compared to single-source CBCT systems. The reconstructed images by DT-FDK also show improved image quality compared to traditional reconstruction algorithms.
Coral ecosystem not only breeds abundant organisms, but also deem to be a very important fishery and tourism resources. The reduction of coral will also have a bad impact on the marine. Scattering and absorption leads...
详细信息
Upright position CT scans make it possible for full-length-body imaging at conditions more relevant to daily situations, but the substantial weight of the upright CT scanners increases the risks to floor’s stability ...
详细信息
ISBN:
(数字)9798350313338
ISBN:
(纸本)9798350313345
Upright position CT scans make it possible for full-length-body imaging at conditions more relevant to daily situations, but the substantial weight of the upright CT scanners increases the risks to floor’s stability and patients’ safety. Robotic-arm CBCT systems are supposed to be a better solution for this task, but such systems still face challenges including long scanning time and low reconstruction quality. To address the above challenges, this paper proposes a novel method to calculate optimal scanning pitch based on data completeness analysis, which can complete the whole-body scan in the shortest time without a significant decline in image quality. Besides, an FDK-style reconstruction method based on normalized projections is proposed to obtain fast image reconstruction. Extensive experiments prove the effectiveness of the proposed optimal scanning trajectory. Qualitative and quantitative comparisons with FDK and iterative algorithms show that the proposed reconstruction method can obtain high imaging quality with reasonable computation costs. The method proposed in this paper is expected to promote the application of robotic-arm CBCT systems in orthopedic functional analysis.
Due to the critical importance of underwater pipeline integrity, particularly in the oil and gas transportation sector. This paper addresses the significance of applying low-rank matrix and sparse representation theor...
Due to the critical importance of underwater pipeline integrity, particularly in the oil and gas transportation sector. This paper addresses the significance of applying low-rank matrix and sparse representation theories in the context of underwater imaging, with a primary focus on oil pipeline inspection. By comparing the effects of three algorithms on imageprocessing, the superiority of the Structured Low-Dimensional Representation (SLDR) algorithm is demonstrated. The research aims to enhance the accuracy and quality of underwater images, ensuring the safety and integrity of subaqueous pipeline systems.
Deep learning (DL) systems have exhibited remarkable capabilities in various domains, such as image classification, natural language processing, and recommender systems, thereby establishing themselves as significant ...
详细信息
Snapshot compressive imaging (SCI) recovers high-dimensional (3D) data cubes from a single 2D measurement, enabling diverse applications like video and hyper-spectral imaging to go beyond standard techniques in terms ...
ISBN:
(纸本)9798331314385
Snapshot compressive imaging (SCI) recovers high-dimensional (3D) data cubes from a single 2D measurement, enabling diverse applications like video and hyper-spectral imaging to go beyond standard techniques in terms of acquisition speed and efficiency. In this paper, we focus on SCI recovery algorithms that employ untrained neural networks (UNNs), such as deep image prior (DIP), to model source structure. Such UNN-based methods are appealing as they have the potential of avoiding the computationally intensive retraining required for different source models and different measurement scenarios. We first develop a theoretical framework for characterizing the performance of such UNN-based methods. The theoretical framework, on the one hand, enables us to optimize the parameters of data-modulating masks, and on the other hand, provides a fundamental connection between the number of data frames that can be recovered from a single measurement to the parameters of the untrained NN. We also employ the recently proposed bagged-deep-image-prior (bagged-DIP) idea to develop SCI Bagged Deep Video Prior (SCI-BDVP) algorithms that address the common challenges faced by standard UNN solutions. Our experimental results show that in video SCI our proposed solution achieves state-of-the-art among UNN methods, and in the case of noisy measurements, it even outperforms supervised solutions. Code is publicly available at https://***/Computational-Imaging-RU/SCI-BDVP.
With the escalating demand for image observation and video transmission in underwater environments, real-time wireless multimedia transmission stands as the pivotal stride in the advancement of ocean exploration and o...
详细信息
ISBN:
(数字)9798350355895
ISBN:
(纸本)9798350355901
With the escalating demand for image observation and video transmission in underwater environments, real-time wireless multimedia transmission stands as the pivotal stride in the advancement of ocean exploration and observation systems. While acoustic communication remains the most dependable solution for underwater wireless communication, multimedia transmission still grapples with the challenges of limited band-width and poor channel quality. In this paper, we introduce a self-developed underwater acoustic modem especially tailored to real-time multimedia transmission. This comprehensive design spans across hardware components, the application layer, data link layer, and the physical layer. The utilization of multi-band retransmission mechanisms and sophisticated physical layer algorithms contributes to the system's resilience against the complexities of the underwater acoustic channel. Subsequent communication experiments have been conducted in both Zhoushan, Zhejiang, and the South China Sea, focusing on horizontal and vertical underwater acoustic channels respectively. Through these trials, the system's demodulation and retransmission capabilities have been duly validated, resulting in the successful transmission of multimedia video files on multiple occasions.
Total focusing method (TFM), as the most effective post-processing technique for ultrasonic phased array imaging currently, can achieve focusing at any point within the detection area, possessing advantages of high re...
详细信息
ISBN:
(数字)9798331516550
ISBN:
(纸本)9798331516567
Total focusing method (TFM), as the most effective post-processing technique for ultrasonic phased array imaging currently, can achieve focusing at any point within the detection area, possessing advantages of high resolution and signal-to-noise ratio that traditional phased array imaging methods lack. However, Total Focusing Method (TFM) imaging requires delay-and-sum processing of full matrix capture (FMC) data. If the number of elements involved in signal acquisition is N, the full matrix data will contain N A-scan signals, each containing thousands of sampling *** will result in long computation time and make it difficult to achieve real-time imaging. This paper proposes a fast total focusing method that combines elliptical arc scanning algorithm and OpenCL hardware acceleration, and deploys it to a self-developed ultrasound phased array imaging system. Results show that the fast total focusing method can generate a 512 × 512 pixel image with 32 elements in an average time of only 64 milliseconds, far surpassing the time required by conventional total focusing algorithms, providing a feasible approach for real-time high-precision non-destructive testing online.
暂无评论