检索结果-内蒙古大学图书馆

Artificial Intelligence and Machine Vision (AIMV), International conference on

作者： Harsh Rakesh Patel Jash Tejaskumar Doshi Courant Institute of Mathematics New York University New York USA

ISBN: (纸本)9781665442121

image processing and action recognition in images are one of the most researched topics in Deep learning. Combining these two concepts for action recognition in lowlight footage is useful in a variety of applications, including night surveillance and self-driving at night. Due to the low photon count and SNR, video in low light is difficult. Short exposures videos are prone to noise, while long exposures can result in blur and are often impractical. To get a better understanding of the presented Action Recognition in Dark(ARID) dataset, which has low light videos divided into it’s action, making it an image classification problem. We examined it in depth and demonstrated it’s utility using simulated dark images. On this dataset, we also benchmarked the performance of existing action recognition models and investigated possible strategies for improving their performance. We introduce a novel pipeline for low-light images using RenNets and statistical image processing methods to identify the human’s actions in it to support the development of learning-based pipelines for human actions recognition in dark videos. We present promising findings from the latest dataset improving the top-1 accuracy by 3.8%. We also examined performance-related causes, and identify areas for potential research.

关键词： Training Deep learning Visualization image recognition Surveillance Machine vision Pipelines

来源：评论

学校读者我要写书评

暂无评论

Detection of Weather Signal in Ground Clutter for Airborne Weather Radar Based on Spatial processing in Elevation 5

Detection of Weather Signal in Ground Clutter for Airborne W...

引用

5th IEEE International conference on Signal and image processing, ICSIP 2020

作者： Wang, Yu Wu, Di Zhu, Daiyin Meng, Fanwang Nanjing University of Aeronautics and Astronautics Key Laboratory of Radar Imaging and Microwave Photonics Ministry of Education Nanjing China Avic Leihua Electronic Technology Research Institute Wuxi China

ISBN: (纸本)9781728168968

An efficient method to achieve weather signal detection (WSD) for airborne weather radar is to use the elevation information collected by vertical displayed multi-channel system. In this paper, we detect the weather signal component contaminated by complex Gaussian ground clutter over a so-called interferometric phase diagram (InPD) generated by dual-channel data. The value in each cell of the InPD, i.e., the interferometric phase, is employed as the test statistic for WSD. The statistics of the interferometric phase crossing elevation channels has been analyzed in detail. Based on the statistical characteristics, a constant false-alarm rate (CFAR) detector is designed for WSD. As demonstrated by simulated and experimental results, the proposed algorithm overperforms most of the current WSD methods. © 2020 IEEE.

关键词： Signal detection

来源：评论

学校读者我要写书评

暂无评论

A hybrid image processing algorithm to examine diseases in citrus leaves and fruits

引用

AIP conference Proceedings 2024年第1期3122卷

作者： P. Usha J. Vijayakumar P. Nisha Department of Electronics and Instrumentation Bharathiar University Coimbatore 641046 India

Agricultural productivity is problematic while the plant attacks several micro-organisms, viruses, and bacterial infections. Earlier disease identification is a unity regarding the challenging solutions for increasing plant production. The signs from these attacks are generally identified in the leaves, fruit, and stems inspection. Nowadays, the need for computational diagnosis of plant diseases increases to gain efficient plant productivity. Here, the process identifies and analyzes the Citrus leaf and Fruits disease literally from the infected images by adopting image-processing procedures to distinguish leaf infections from digital-images. The suggested hybrid algorithm includes pre-processing and RGB HSI conversion. The initial step practices CLAHE segments the features of affected areas applying K-means clustering and statistical GLCM. The SVM classifier has recognized the diseased image and performs those methods in citrus disease detection. Finally, the Fuzzy-based estimation has been convoluted to measure the disease grade severity.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Proceedings of the 23rd International conference on Digital Audio Effects, eDAFx2020 and DAFx2020 in 21

Proceedings of the 23rd International Conference on Digital ...

引用

23rd International conference on Digital Audio Effects, eDAFx 2020 and DAFx 2021

The proceedings contain 43 papers. The topics discussed include: room impulse response estimation using signed distance functions;an acoustic paintbrush method for simulated spatial room impulse responses;flexible real-time reverberation synthesis with accurate parameter control;evaluation of a stochastic reverberation model based on the source image principle;identification of nonlinear circuits as port-Hamiltonian systems;an equivalent circuit interpretation of antiderivative antialiasing;non-iterative schemes for the simulation of nonlinear audio circuits;applications of port Hamiltonian methods to non-iterative stable simulations of the KORG35 and MOOG 4-pole VCF;deforming the oscillator: iterative phases over parametrizable closed paths;continuous state modeling for statistical spectral synthesis;subjective evaluation of sound quality and control of drum synthesis with Stylewavegan;and a low-latency quasi-linear-phase octave graphic equalizer.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Soil organic matter estimation by using Landsat-8 pansharpened image and machine learning 4

Soil organic matter estimation by using Landsat-8 pansharpen...

引用

4th International conference on Intelligent Computing in Data Sciences (ICDS)

作者： Bouasria, Abdelkrim Namr, Khalid Ibno Ettachfini, El Mostafa Rahimi, Abdelmejid Chouaib Doukkali Univ Fac Sci Dept Geol El Jadida Morocco

ISBN: (纸本)9781728180847

Considering the significant position of soil organic matter (SOM) in soil quality and maintenance, and its role in the functioning of soil physicochemical and biological processes, it is essential to monitor frequently the SOM status and its dynamics. It is a time-consuming and expensive task if we depend exclusively on chemical analysis, particularly in a semi-arid irrigated zone and with intensive agricultural activities and a very fragmented landscape. It is the Sidi Bennour region, which is situated in Doukkala Irrigated Perimeter in Morocco. Data from satellites could be a good alternative to conventional methods and fill this void with low costs. There has been a great deal of interest in satellite image prediction models, especially with free and abundant availability of satellite data. This work intends to predict SOM using Decision Trees (DT), k-Nearest Neighbors (k-NN), and Artificial Neural Networks (ANN). The soil samples (369 points) were collected at 0-30 cm of depth and the laboratory analysis was carried out. A multispectral Landsat-8 image was acquired and then calibrated. An image pansharpening processing was applied to produce a PAN image with 15m of resolution from 30m image resolution (MS). The obtained results indicate that the ANN model outperformed the other predictive models for both images (MS and PAN) with R-2 = 0.6553 and R-2 = 0.6985, respectively. The statistical RMSE of predictive models was 0.2153 and 0.2014, and MAE was 0.1682 and 0.1573 for both images, MS and PAN respectively. For this predictive model, the image pansharpening could increase the prediction accuracy of R-2 by 4.32% and reduce the RMSE by 1.39%.

关键词： Soil organic matter digital soil mapping remote sensing machine learning neural networks

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Stereo-Matching Technologies for Emphasizing Stereo-Vision Applications

Comparative Analysis of Stereo-Matching Technologies for Emp...

引用

Advances in Computing, Communication Control and Networking (ICAC3N), International conference on

作者： V. Kavitha Swami Dayananda College of Arts & Science Manjakudi India

ISBN: (纸本)9781665438124

Stereo vision is a growing research domain which seeks the attention of various researchers to attain deeper scene extraction. This work provides an extensive analysis towards the stereo-matching algorithms and stereo-vision to resolve the problems related to it. The analysis towards the stereo matching technologies is executed with benchmark standards with the focus on stereo vision methods. Thus the comparison of stereo matching algorithms can be done through the implementation of stereo vision application in a particular domain so that the results for the algorithms are comparatively analyzed. In most cases, the analysis and comparison are performed with statistical analysis and emphasize the benefits of various stereo algorithms. Some approaches give higher computational cost with expected outcomes over the lower time frame and provides competency towards parallel processing. The results obtained from the various stereo matching algorithms through the identified different parameters of bad pixels.

关键词： Costs Target tracking image resolution statistical analysis Navigation Process control Parallel processing

来源：评论

学校读者我要写书评

暂无评论

Leukocytes Segmentation and Classification in Digital Microscopic images

Leukocytes Segmentation and Classification in Digital Micros...

引用

Computing & Information Sciences (ICCIS), International conference on

作者： Muhammad Abbas Hussain Ibtihaj Ahmad Arslan Shaukat Zain UI Islam College of EME National University of Sciences and Technology (NUST) Pakistan Foundation University Islamabad (FUI) Pakistan

ISBN: (纸本)9781665494427

image processing and machine learning have recently gained positive contributions to various medical procedures. One of the diagnostic processes' essential requirements in many diseases is laboratory tests, such as the Complete Blood Count (CBC) test. In CBC, various leukocytes, also known as White Blood Cells (WBC), are segmented, classified, and counted by a lab technician in microscopic slides. This process is very tiresome and requires a human technician with specialized skill sets. This research proposes a fully automatic algorithm for the segmentation and classification of white blood cells. The proposed method applies pre-processing techniques to digital microscopic images. White blood cells are then segmented based on color pallets. Hybrid features are extracted from the segmented images based on the fusion of local binary patterns and statistical features. Then various classifiers are used for the classification of WBC. Results suggest that the Support Vector Machine (SVM) and Artificial Neural Networks (ANN) outclass other classifiers. It was observed that the proposed methodology outperformed existing methods in terms of classification accuracy (97.5%).

关键词： White blood cells Support vector machines image segmentation image color analysis Microscopy Artificial neural networks Cells (biology)

来源：评论

学校读者我要写书评

暂无评论

New Born Face Feature Extraction Recognition using image processing Techniques

New Born Face Feature Extraction Recognition using Image Pro...

引用

Interdisciplinary Humanitarian conference for Sustainability (IIHC), International

作者： Shashikala H. K Gadug Sudhamsu Department of Computer Science and Engineering Faculty of Engineering and Technology Jain (Deemed-to-be University) Bangalore

Biometric techniques such as fingerprints, palm prints, irises, faces, and retinas are used to identify the individuals responsible for the forgery. The importance of facial recognition has grown as a direct result of the rise in the number of fakes. The information that makes up the repository for newborn babies comes from a wide variety of sources, and it features babies in a variety of poses and lighting conditions throughout many distinct galleries. A number of different facial recognition strategies and classification algorithms currently used in commercial systems are also examined as part of a standard benchmark test. The data set is obtained from the face of a newborn baby, then trained and tested with classification algorithms, and then compared based on a variety of performance indicators. The extractable features from a picture serve as the basis for the classification process, and some of these extracted features are also used for the purposes of training and testing in conjunction with the classification process. Different feature extraction strategies are investigated in this study, including local binary patterns (LBP), principal component analysis (PCA), and gray level co-occurrence matrix (GLCM). LBPs are described by their local binary patterns. Eigenfaces and Eigenvectors are produced using Principal Component Analysis, and second-order statistical features are constructed with Gray Level Co-occurrence Matrix. Photographs of newborn babies displaying a variety of facial expressions use these techniques. The extracted features are provided as input to the support vector machine for classification. In comparison to the other feature extraction methods, the principal component analysis method had an accuracy of 91%, a better recognition rate, and a shorter computation time. Experiments have also shown that this method is superior to other.

关键词： Training Support vector machines Pediatrics Hospitals Face recognition Feature extraction Classification algorithms

来源：评论

学校读者我要写书评

暂无评论

2nd International conference on Machine Intelligence and Signal processing, MISP 2019

2nd International Conference on Machine Intelligence and Sig...

引用

2nd International conference on Machine Intelligence and Signal processing, MISP 2019

ISBN: (纸本)9789811513657

The proceedings contain 36 papers. The special focus in this conference is on Machine Intelligence and Signal processing. The topics include: Real-time RADAR and LIDAR sensor fusion for automated driving;generalizing streaming pipeline design for big data;Adaptive fast composite splitting algorithm for MR image reconstruction;extraction of technical and non-technical skills for optimal project-team allocation;modified flower pollination algorithm for optimal power flow in transmission congestion;Intelligent condition monitoring of a CI engine using machine learning and artificial neural networks;bacterial foraging optimization in non-identical parallel batch processing machines;healthcare information retrieval based on neutrosophic logic;Convolutional neural network long short-term memory (CNN + LSTM) for histopathology cancer image classification;a novel approach for music recommendation system using matrix factorization technique;forecasting with multivariate fuzzy time series: A statistical approach;nature-inspired algorithm-based feature optimization for epilepsy detection;a combined machine-learning approach for accurate screening and early detection of chronic kidney disease;backpropagation and self-organizing map neural network methods for identifying types of eggplant fruit;head pose prediction while tracking lost in a head-mounted display;recommendation to group of users using the relevance concept;ACA: Attention-based context-aware answer selection system;dense and partial correspondence in non-parametric scene parsing;audio surveillance system;mopsa: Multiple output prediction for scalability and accuracy;Generation of image captions using VGG and resnet CNN models cascaded with RNN approach;impact of cluster sampling on the classification of landsat 8 remote sensing imagery;deep neural networks for out-of-sample classification of nonlinear manifolds;FPGA implementation of LDPC decoder.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Training stochastic stabilized supralinear networks by dynamics-neutral growth 22

Training stochastic stabilized supralinear networks by dynam...

引用

Proceedings of the 36th International conference on Neural Information processing Systems

作者： Wayne W.M. Soo Máté Lengyel Department of Engineering University of Cambridge Department of Engineering University of Cambridge and Department of Cognitive Science Central European University

ISBN: (纸本)9781713871088

There continues to be a trade-off between the biological realism and performance of neural networks. Contemporary deep learning techniques allow neural networks to be trained to perform challenging computations at (near) human-level, but these networks typically violate key biological constraints. More detailed models of biological neural networks can incorporate many of these constraints but typically suffer from subpar performance and trainability. Here, we narrow this gap by developing an effective method for training a canonical model of cortical neural circuits, the stabilized supralinear network (SSN), that in previous work had to be constructed manually or trained with undue constraints. SSNs are particularly challenging to train for the same reasons that make them biologically realistic: they are characterized by strongly-connected excitatory cells and expansive firing rate non-linearities that together make them prone to dynamical instabilities unless stabilized by appropriately tuned recurrent inhibition. Our method avoids such instabilities by initializing a small network and gradually increasing network size via the dynamics-neutral addition of neurons during training. We first show how SSNs can be trained to perform typical machine learning tasks by training an SSN on MNIST classification. We then demonstrate the effectiveness of our method by training an SSN on the challenging task of performing amortized Markov chain Monte Carlo-based inference under a Gaussian scale mixture generative model of natural image patches with a rich and diverse set of basis functions - something that was not possible with previous methods. These results open the way to training realistic cortical-like neural networks on challenging tasks at scale.

关键词：

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：