检索结果-内蒙古大学图书馆

Autonomous Obstacle Avoidance Scheme Using Monocular Vision Applied to Mobile Robots

Autonomous Obstacle Avoidance Scheme Using Monocular Vision ...

2020 International Conference on intelligent Control, Measurement and Signal Processing and intelligent Oil Field, ICMSP 2020

作者： Song, Yun Yun Yang, Sheng Yi Liu, Chao Wu, Qing Gui School of Data Science and Information Engineering Key Laboratory of Pattern Recognition and Intelligent Systems of Guizhou Province Guizhou Minzu University Guiyang China School of Mechatronics Engineering Guizhou Minzu University Guiyang China

Mobile robots cannot move in an unknown environment with static or slow-moving obstacles effectively. We present an enhanced obstacle avoidance strategy using monocular vision to solve this problem. First, we combine Canny and Otsu to extract the barrier feature and find the critical pixel position of obstacles by the monocular vision. Then the image depth estimation algorithm is used to estimate the gaps. With these parameters of barriers, an improved bug algorithm is proposed to avoid the obstacles autonomously. The experiments show that the proposed obstacles avoidance strategy can effectively make a small mobile robot avoid different kinds of obstacles. © Published under licence by IOP Publishing Ltd.

关键词： Mobile robots

来源：评论

学校读者我要写书评

暂无评论

Cd-Vae: An Unsupervised Disentangled Representation Learning Framework for Visual Data

SSRN

引用

SSRN 2023年

作者： Huang, Chengquan Cai, Jianghai Luo, Senyan Wang, Shunxia Yang, Guiyan Lei, Huan Zhou, Lihua Guizhou Key Laboratory of Pattern Recognition and Intelligent System Guizhou Minzu University Guiyang550025 China School of Data Science and Information Engineering Guizhou Minzu University Guiyang550025 China

Disentangled representation learning is a crucial research problem in the field of artificial intelligence, and learning meaningful representation of visual data is useful for improving the generalizability and interpretability of machine learning models. To discover and learn the interpretable factors behind the visual data, previous approaches typically use extra regularization terms in learning disentangled representations of the generated images, which usually lead to poor results between disentanglement and generative quality. For the generative model variational auto-encoder (VAE), the traversal of the generated images along different factor directions shows meaningful and interpretable variations in the latent space. Building upon this observation, we argue that it is important to learn disentangled representations of the generated images by exploiting the scalability and training stability of the generative model VAE, and focusing on meaningful traversal directions in the latent space. Therefore, we propose contrast disentanglement based on VAE (CD-VAE), a deep unsupervised disentangled representation learning framework for visual data, that can discover the latent factors in image data. CD-VAE jointly learns disentangled representations by traversing image variations in the latent space, and constructing contrast disentanglement in the variation space. Experimental results demonstrate that CD-VAE can discover interpretable directions and achieve better disentangled representation learning for visual data. Source codes and pre-training data will be publicly available at GitHub website. © 2023, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Fire Detection Based on Flame Enhancement for Weak Fires 19th

Fire Detection Based on Flame Enhancement for Weak Fires

引用

19th Chinese Conference on Image and Graphics Technologies and Applications, IGTA 2024

作者： Chen, Kuan Wen, Wen Feng, Fujian Xu, Xiang Liang, Yihui School of Computer Guangdong University of Technology Guangzhou510000 China School of Computer Science Zhongshan Institute University of Electronic Science and Technology of China Zhongshan528400 China Guizhou Key Laboratory of Pattern Recognition and Intelligent System Guizhou Minzu University Guiyang550025 China

ISBN: (纸本)9789819799183

Detecting weak fire, such as overexposed and highly transparent flames, remains a significant challenge in vision-based fire detection. Convolutional Neural Network (CNN) based methods are widely used for automatic fire feature extraction, but they struggle to accurately recognize overexposed flames similar to the background and highly transparent flames that blend with the background, leading to false negative fire detection results. To address this issue, we have established a large-scale fire detection dataset with weak flames and introduce a fire detection method based on flame context enhancement. While employing a flame feature extraction module to identify and delineate the potential flame area, the proposed method primarily introduces a flame context module to capture the effective features of weak flames from this area, significantly enhancing the understanding of fires. Experimental results show that the proposed method can effectively reduce false negatives and improves the recall rate for detecting weak flames. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Premixed flames

来源：评论

学校读者我要写书评

暂无评论

Adaptive Pixel Pair Generation Strategy for Image Matting Methods Based on Pixel Pair Optimization 19th

Adaptive Pixel Pair Generation Strategy for Image Matting Me...

引用

19th Chinese Conference on Image and Graphics Technologies and Applications, IGTA 2024

作者： Zheng, Jiamin Wen, Wen Liang, Yihui Feng, Fujian Xu, Xiang School of Computer Guangdong University of Technology Guangzhou510000 China School of Computer Science Zhongshan Institute University of Electronic Science and Technology of China Zhongshan528400 China Guizhou Key Laboratory of Pattern Recognition and Intelligent System Guizhou Minzu University Guiyang550025 China

ISBN: (纸本)9789819799183

Natural image matting plays a crucial role in numerous real-world applications. Image matting methods based on pixel pair optimization is a type of matting algorithm, which has significant advantages in parallel computation, handling images with similar foreground and background, and operating under limited computing time. However, these methods cannot provide high-quality alpha mattes for high-resolution images within limited computational resources. In this paper, we design an adaptive pixel pair generation strategy to solve the above problem. This strategy adaptively generates pixel pair for the unknown pixel according to its estimated alpha value, which promotes diversity of pixel pair and improves the efficiency of pixel pair optimization. It categorizes unknown pixels into three types and additionally incorporates nearby foreground and background pixels to generate pixel pair. Experimental results show that using our strategy can achieve high-quality alpha mattes and competitive matting performance under limited computational resources. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

Feature Decoupled of Deep Mutual Information Maximization

Feature Decoupled of Deep Mutual Information Maximization

引用

Automation, Robotics and Computer Engineering (ICARCE), International Conference on

作者： Xing He Changgen Peng Lin Wang Weijie Tan Zifan Wang State Key Laboratory of Public Big Data College of Computer Science and Technology Guizhou University Guiyang China Guizhou Big Data Academy Guizhou University Guiyang China Guizhou Key Laboratory of Pattern Recognition and Intelligent System Guizhou Minzu University Guiyang China Institute of Guizhou Aerospace Measuring and Testing Technology Guiyang China

In deep learning, supervised learning techniques usually require a large amount of expensive labeled data to train the network, and the feature representations extracted by the model usually mix multiple attributes, resulting in feature representations that are difficult to decouple and are non-interpretable, which restricts the application and development of deep learning techniques, and for this reason, it is particularly important to study decoupled feature representation methods for unsupervised learning. Although the Learning deep representations by mutual information estimation and maximization (DIM) method achieves excellent results in unsupervised learning, the feature representations learned by the DIM method still suffer from the problem of difficult decoupling. decoupling problem. To address this problem, we minimize the mutual information between the intermediate layer feature representations learned by the hidden layer of the encoder during the encoder training process, so that the features learned by each filter are as uncorrelated as possible, thus realizing feature decoupling, and our method is called FP-DIM. Finally, the effectiveness of the FP-DIM method is verified on the CIFAR-10, STL-10, and Fashion-MNIST datasets. The experiments show that our proposed FP-DIM method is more significant for the learned decouplable middle layer feature representation. Finally, we also propose a reflection of future research for the FP-DIM method, aiming to provide a research idea and direction for solving unsupervised interpretable machine learning and to lay a solid theoretical and application foundation for machine learning fields such as image classification and migration learning.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Mixture Loss Function-based Classification Network for Few-shot Learning

Mixture Loss Function-based Classification Network for Few-s...

引用

Computing, Robotics and System Sciences (ICRSS), International Conference on

作者： Yansha Zhang Feng Pan Jie Wang Lin Wang College of Date Science and Information Engineering GuiZhou Minzu University Guiyang China Key Laboratory of Pattern Recognition and Intelligent System of Guizhou Province Guiyang China

Data augmentation technology can effectively solve the problem of few-shot image classification, but many approaches based on data augmentation generated sample have poor discriminability, which negatively affects the classifier and classification accuracy needs to be improved. In this regard, based on the TFFCN, a mixture loss function-based classification network for few-shot learning (MLFCN) is proposed. First of all, for the prototype tensor features with a simple mean representation, ignoring the different contributions of the sample features of support set for per class to the class prototype tensor features, so that the generated prototype tensor features can not well represent sample features of per class, affecting the discriminability of the generated samples. Introduce the improved adaptively spatial feature fusion module, it uses the convolution, considers the support set of different sample features on the prototype tensor, adaptively assign weights, then weighted and summed to obtain a more representative prototype tensor feature; secondly, aiming at the blurred classification boundary of similar categories in the generated sample features, a mixture loss function is proposed. The loss function is to add the inter-class mean squared error term to the mean square error loss, and stretches the inter-class distance of the generated tensor feature, so that the classification boundary is clearer, so as to train a better classifier. Experimental results show that the improved adaptively spatial feature fusion and the mixing loss function can augment the discriminability of the generated tensor features and further improve the classification accuracy.

关键词： Training Tensors Convolution Prototypes Mean square error methods Feature extraction Generators

来源：评论

学校读者我要写书评

暂无评论

Stable and Feature Decoupled of Deep Mutual Information Maximization Based on Wasserstein Distance

SSRN

引用

SSRN 2023年

作者： He, Xing Peng, Changgen Wang, Lin Tan, Weijie State Key Laboratory of Public Big Data College of Computer Science and Technology Guizhou University PR Guiyang550025 China Guizhou Key Laboratory of Pattern Recognition and Intelligent System Guizhou Minzu University PR Guiyang550025 China Guizhou Big Data Academy Guizhou University PR Guiyang550025 China Key Laboratory of Advanced Manufacturing Technology Ministry of Education Guizhou University PR Guiyang550025 China

Deep learning techniques usually require large amounts of expensive labeled data to train networks, and the extracted deep representations are usually mixed with multiple attributes having uninterpretability, which limits the application and development of deep learning techniques. Therefore, it is crucial to investigate the stability of training models by unsupervised learning methods and to learn feature decoupled representations. Although the method based on depth mutual information maximization(Deep InfoMax or DIM) has achieved good results in the field of computer vision, it is extremely prone to "gradient explosion" or "gradient disappearance", which makes the training model unstable. At the same time, it is difficult to decouple the features learned by the encoder. In order to solve these problems, we stable and feature decoupled of deep mutual information maximization based on Wasserstein distance, which we call the method called WDIM. The method first addresses the instability of the model trained by the DIM method and proposes to measure the distance between the encoder output and the prior distribution based on the Wasserstein distance, which is a good guide to the stability of the model training. Second, to address the problem that the features learned by the encoder are difficult to be decoupled, the mutual information between the features learned by the hidden layer of the encoder and the intermediate layer is required to be minimized during the training of the encoder so that the features learned by each filter are as uncorrelated as possible to achieve feature decoupling. Finally, the WDIM method is validated on the CIFAR-10, CIFAR-100, STL-10, and FashionMNIST datasets. Experiments show that our proposed WDIM method is more stable for model training, faster model convergence, and more significant feature decoupling. © 2023, The Authors. All rights reserved.

关键词： Network coding

来源：评论

学校读者我要写书评

暂无评论

Natural image matting based on image inpainting

Natural image matting based on image inpainting

引用

Computer Graphics, Image and Virtualization (ICCGIV), International Conference on

作者： Yuan Zhang Mian Tan Zhulian Zhou Yuan Yang Yihui Liang Fujian Feng Guizhou Key Laboratory of Pattern Recognition and Intelligent System Guizhou Minzu University Guiyang China School of Computer Science University of Electronic Science and Technology China Zhongshan China

Deep image matting is a hot problem with applications in computer vision and image processing. It has been widely used in image composition, film production and video editing etc. The current matting method based on image inpainting uses a deep neural network to inpaint the foreground target and background region to further improve the accuracy of alpha matte. However, when the trimap contains large unknown regions, the excessive inpainting of foreground and background produce a lot of redundant information, which leads to a degradation of alpha matte quality. Therefore, to address this issue, a natural image matting method based on image inpainting is designed. This method involves the refinement process for trimap, which improves the quality of the trimap, enlarges the foreground and background regions providing additional information for image matting. Extensive experimental results on the composition-1k dataset demonstrate that the presented method provide high-quality alpha mattes not only in the case that the trimap contains small unknown regions, but also in the case that the trimap contains large unknown regions.

关键词： Degradation Deep learning Computer vision Image processing Refining Neural networks Production

来源：评论

学校读者我要写书评

暂无评论

Tight reservoirs classification using random forest: A case study of he 8 member in eastern yan'an gas field 3

Tight reservoirs classification using random forest: A case ...

引用

3rd International Conference on intelligent Control, Measurement and Signal Processing and intelligent Oil Field, ICMSP 2021

作者： Yan, Wang Ruogu, Wang Shengyi, Yang Jianping, Liu College of Petroleum Engineering Xi'An Shiyou University Xi'an China Co. Ltd. Xi'an China Guizhou Minzu University Key Laboratory of Pattern Recognition and Intelligent Systems of Guizhou Province Guiyang China Changqing Oilfield Company Exploration Department of PetroChina Xi'an China

ISBN: (纸本)9781665437158

Tight sandstone reservoir is very important in oil and gas exploration in China. Tight reservoirs classification and evaluation are a frontier research field. There are many indexes involved in reservoirs classification, and it is necessary to judge the reservoir type according to personal experience, which consumes lots of time and manpower. Therefore, a new classification method of tight reservoirs using random forest is proposed. Firstly, the high pressure mercury injection curves of tight sandstone reservoirs of He 8 member of Lower Shihezi Formation in eastern Yan'an Gas Field are selected as the research data. Four characteristics for classification are obtained by principal component analysis. Secondly, the random forest using CART is used to classify and obtain the results of reservoir classification. Finally, classification results are verified and parameters of the random forest are optimized. Experimental results show that the proposed reservoirs classification method has high accuracy and low calculation cost. It can effectively reduce time loss and save manpower, and has good generalization. © 2021 IEEE.

关键词： Principal component analysis

来源：评论

学校读者我要写书评

暂无评论

Geometric Numerical Integral Method in Compact Lie Group

Geometric Numerical Integral Method in Compact Lie Group

引用

6th International Conference on intelligent Computing and Signal Processing (ICSP)

作者： Chao Liu Shengyi Yang Key Laboratory of Pattern Recognition and Intelligent System of Guizhou Province Guizhou Minzu University Gui Yang China

ISBN: (纸本)9781665447300

Three dimensions special orthogonal group SO (3) is widely used to describe the rotation kinematics of the rigid body without local coordinates, which can avoid rotation singularity and unwinding in traditional methods. Propagating the rotation kinematics in SO (3) with a specific geometric integration method is not only to obtained numerical results with improved qualitative behavior, but also provided more accurate long-time integration results. While many studies have focused on geometric integration algorithms to preserve the geometric structure, this work has the additional objective of studying result accuracy. Integral curves on SO (3) obtained using the third-order Crouch-Grossman Lie group method are compared with numerical results using the third-order RKMK algorithms, the exponential coordinates method, and the simple projection method. Results show that the use of the Crouch-Grossman Lie group method better preserves the geometric structure of SO (3) for the larger time steps considered. It is also found that the third-order Crouch-Grossman algorithm is more accurate than the RKMK except for the smallest time step used.

关键词： Heuristic algorithms Signal processing algorithms Kinematics Signal processing

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：