检索结果-内蒙古大学图书馆

Building an image set for modeling image re-targeting using deep learning

SIMULATION MODELLING PRACTICE AND THEORY 2023年第1期126卷

作者： Alsmirat, Mohammad A. Qawasmeh, Ethar Al-Ayyoub, Mahmoud Jararweh, Yaser Univ Sharjah Dept Comp Sci Sharjah U Arab Emirates Jordan Univ Sci & Technol Dept Comp Sci Amman Jordan Ajman Univ Coll Engn & Informat Technol Artificial Intelligence Res Ctr AIRC Ajman U Arab Emirates

The displays of various devices such as TVs, laptops, and mobile phones vary greatly in size, aspect ratio, and resolution. Creating different versions of the same media for these different systems can be costly in terms of storage. A novel solution to this problem is image re-targeting, which adapts an image to fit any device display, preserving the main content of the image without wasting additional storage. Re-targeting can be performed on the device itself or on an intermediate device such as an edge server. The main re-targeting techniques found in literature are cropping, seam carving, and scale and stretch, but it is not easy to determine the best technique for a given image. In this paper, we propose using deep learning to model the decision of which re-targeting technique to use for which target size. To achieve this, we built and annotated a large image dataset using a team of 10 people, including 2750 original images representing 6 categories. We applied the most commonly used four re-targeting techniques (cropping, scaling, Seam Carving, and optimized scale and stretch) to the original image set to create five different sizes (HDP (1600 x 900), HD (1280 x 720), SVGA (800 x 600), WVGA (852 x 480), and NTSC (720 x 480)). The resulting image set, which includes the re-targeted images, contains 46750 images. We examined and summarized the annotators' perceptual point of view for reference and comparison purposes. Additionally, we created a deep learning model and trained it using the newly generated image set to recommend the best re-targeting technique based on an input image and a target size. The model consists of 30 independent sub-models (five models for each image category). All these sub-models are constructed using pre-trained ResNet50. The choice of the sub-model to use for a particular image is based on both the input image category and target size. To accelerate the training process and avoid overfitting of the sub-models, we utilize transfer lea

关键词： image re-targeting image datasets QoE Human perceptual views Deep learning Transfer learning

来源：评论

学校读者我要写书评

暂无评论

A survey on analysis of human faces and facial expressions datasets

引用

INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS 2020年第3期11卷 553-571页

作者： Khan, Gulraiz Samyan, Sahar Khan, Muhammad Usman Ghani Shahid, Muhammad Wahla, Samyan Qayyum Univ Engn Al khawarizmi Inst Comp Sci & Technol Lahore Pakistan

Facial expressions are the basic input for visual emotion detection. They are of great importance in computer vision society. In last decade, substantial amount of work has been done in the field of facial expressions datasets. This survey covers all of the publically available databases in detail and provides necessary information about these sets. This review delivers comprehensive support to researchers in selection of their desired dataset. The datasets are organized in decreasing order of their importance with respect to diversity in expressions, poses, number of images and resolution. This survey also provides comprehensive tabular comparison of different face based databases.

关键词： Emotion Face Facial expressions image datasets Video datasets

来源：评论

学校读者我要写书评

暂无评论

image interpolation with adaptive k-nearest neighbours search and random non-linear regression

引用

IET image PROCESSING 2020年第8期14卷 1539-1548页

作者： Zheng, Jieying Song, Wanru Wu, Yahong Liu, Feng Nanjing Univ Posts & Telecommun Jiangsu Key Lab Image Proc & Image Commun 66 Xin Mofan Rd Nanjing 21003 Peoples R China Nanjing Univ Posts & Telecommun Key Lab Broadband Wireless Commun & Sensor Networ Minist Educ 66 Xin Mofan Rd Nanjing 210003 Peoples R China

Learning-based image interpolation methods have been proved to be effective in image interpolation. In this study, the authors propose an accurate image interpolation with adaptive k-nearest neighbour searching and non-linear regression. The proposed method aims to find k-nearest neighbours of the input image patch and use them to learn the non-linear mapping between low-resolution and high-resolution image patches. To be specific, they first divide the training image patches into many subspaces, then they utilise an adaptive robust and precise k nearest neighbour searching scheme with proposed normalised Gaussian similarity to find the k nearest neighbours in the matched subspace. The selected k image patch pairs are then used to learn the non-linear regression model through an extreme learning machine. Furthermore, the proposed interpolation method is a cascade framework that consists of two stages. Stage 2 takes the results of Stage 1 as input to further improve the performance. Extensive experimental results on commonly used test images and image datasets indicate that their proposed algorithm obtains competitive performance against the state-of-the-art methods both in terms of objective evaluation values and the subjective effect of reconstructed images.

关键词： image reconstruction regression analysis image resolution interpolation nearest neighbour methods search problems random processes image patch pairs nonlinear regression model extreme learning machine test images image datasets reconstructed images nearest neighbours search image interpolation methods input image patch nonlinear mapping high-resolution image patches training image patches normalised Gaussian similarity nearest neighbour searching scheme

来源：评论

学校读者我要写书评

暂无评论

Feature selection in image analysis: a survey

引用

ARTIFICIAL INTELLIGENCE REVIEW 2020年第4期53卷 2905-2931页

作者： Bolon-Canedo, Veronica Remeseiro, Beatriz Univ A Coruna Dept Comp Sci CITIC Campus Elvina S-N La Coruna 15071 Spain Univ Oviedo Dept Comp Sci Campus Gijon S-N Gijon 33203 Spain

image analysis is a prolific field of research which has been broadly studied in the last decades, successfully applied to a great number of disciplines. Since the apparition of Big Data, the number of digital images is explosively growing, and a large amount of multimedia data is publicly available. Not only is it necessary to deal with this increasing number of images, but also to know which features extract from them, and feature selection can help in this scenario. The goal of this paper is to survey the most recent feature selection methods developed and/or applied to image analysis, covering the most popular fields such as image classification, image segmentation, etc. Finally, an experimental evaluation on several popular datasets using well-known feature selection methods is presented, bearing in mind that the aim is not to provide the best feature selection method, but to facilitate comparative studies for the research community.

关键词： Feature selection image analysis Pattern recognition High dimensionality image datasets

来源：评论

学校读者我要写书评

暂无评论

Recent evolution of modern datasets for human activity recognition: a deep survey

引用

MULTIMEDIA SYSTEMS 2020年第2期26卷 83-106页

作者： Singh, Roshan Sonawane, Ankur Srivastava, Rajeev Indian Inst Technol BHU Dept CSE Varanasi 221005 Uttar Pradesh India Indian Inst Technol BHU Dept Elect Engn Varanasi 221005 Uttar Pradesh India

Human activity recognition has been a significant goal of computer vision since its inception and has developed considerably in the last years. Recent approaches to this problem increasingly favour the use of data-driven deep learning methods. To facilitate the comparison of these methods, several datasets pertaining to labelled human activity have been created, having great variation in content and methodology. As the field has developed, the datasets used have undergone considerable evolution as well. In this paper, we attempt to classify and describe a variety of datasets for researchers to choose the most suitable benchmark for their domain. For this, we propose a set of characteristics by which datasets may be compared. We also describe the progress in recent years that sets modern datasets apart from those used in the past.

关键词： Video datasets image datasets Visual datasets Modern datasets Activity recognition Action recognition

来源：评论

学校读者我要写书评

暂无评论

image retrieval based on colour and improved NMI texture features

引用

AUTOMATIKA 2019年第4期60卷 491-499页

作者： Du, Anyu Wang, Liejun Qin, Jiwei Xinjiang Univ Sch Informat Sci & Engn Urumqi Peoples R China Xinjiang Univ Sch Software Engn 499 Northwest Rd Urumqi 830046 Peoples R China Shaanxi Normal Univ Sch Educ Xian Shaanxi Peoples R China

This paper proposes an improved method for extracting NMI features. This method uses Particle Swarm Optimization in advance to optimize the two-dimensional maximum class-to-class variance (2OTSU) in advance. Afterwards, the optimized 2OUSU is introduced into the Pulse Coupled Neural Network (PCNN) to automatically obtain the number of iterations of the loop. We use an improved PCNN method to extract the NMI features of the image. For the problem of low accuracy of single feature, this paper proposes a new method of multi-feature fusion based on image retrieval. It uses HSV colour features and texture features, where, the texture feature extraction methods include: Grey Level Co-occurrence Matrix (GLCM), Local Binary Pattern (LBP) and Improved PCNN. The experimental results show that: on the Corel-1k dataset, compared with similar algorithms, the retrieval accuracy of this method is improved by 13.6%;On the AT&T dataset, the retrieval accuracy is improved by 13.4% compared with the similar algorithm;on the FD-XJ dataset, the retrieval accuracy is improved by 17.7% compared with the similar algorithm. Therefore, the proposed algorithm has better retrieval performance and robustness compared with the existing image retrieval algorithms based on multi-feature fusion.

关键词： CBIR normalized moment of inertia PCNN multi-feature fusion image datasets

来源：评论

学校读者我要写书评

暂无评论

CoLeaf-DB: Peruvian coffee leaf images dataset for coffee leaf nutritional deficiencies detection and classification

引用

DATA IN BRIEF 2023年 48卷 109226页

作者： Tuesta-Monteza, Victor A. Mejia-Cabrera, Heber I. Arcila-Diaz, Juan Univ Senor de Sipan Fac Ingn Arquitectura & Urbanismo Chiclayo Peru Univ Lord of Sipan Fac Ingn Arquitectura & Urbanismo Km 5 Carretera Pimentel Pimentel Chiclayo Peru

This article introduces Peruvian coffee leaf datasets known as CATIMOR, CATURRA and BORBON of coffee plantations loJaen province, Cajamarca, Peru. The leaves with nutritional deficiencies were identified by agronomists, using a physical structure the controlled environment was designed and the images were captured with a digital camera. The dataset contains 1006 leaf images grouped according to their nunesium, Manganese, Nitrogen and others). CoLeaf dataset contain images that facilitate training and validation during the utilization of deep learning algorithms for coffee plant leaf nutritional deficiencies recognition and classification. The dataset is publicly and freely available at http://dx. ***/10.17632/brfgw46wzb.1 .(c) 2023 Published by Elsevier Inc. This is an open access article under the CC BY license ( http://***/licenses/by/4.0/ )

关键词： Peruvian coffee image datasets Machine learning Deep learning Deficiencies detection

来源：评论

学校读者我要写书评

暂无评论

Soybean image dataset for classification

引用

DATA IN BRIEF 2023年 48卷 109300页

作者： Lin, Wei Fu, Youhao Xu, Peiquan Liu, Shuo Ma, Daoyi Jiang, Zitian Zang, Siyang Yao, Heyang Su, Qin Nanjing Agr Univ Nanjing Peoples R China Jiangsu Univ Sci & Technol Zhenjiang Peoples R China Tonghua Normal Univ Tonghua Peoples R China Jiangsu Univ Technol Changzhou Peoples R China

This paper presents a dataset with 5513 images of individ-ual soybean seeds, which encompass five categories : ( I ) In-tact, ( II ) Immature, ( m ) Skin-damaged, ( IV ) Spotted, and ( V ) Broken . Furthermore, there are over 10 0 0 images of soybean seeds in each category. Those images of individ-ual soybeans were classified into five categories based on the Standard of Soybean Classification (GB1352-2009) [1]. The soybean images with the seeds in physical touch were cap-tured by an industrial camera. Subsequently, individual soy-bean images (227 x227 pixels) were divided from the soy-bean images (3072 x2048 pixels) using an image-processing algorithm with a segmentation accuracy of over 98%. The dataset can serve to study the classification or quality assess-ment of soybean seeds.& COPY;2023 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY license ( http://***/licenses/by/4.0/ )

关键词： image processing image datasets Soybean Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Scraping Social Media Photos Posted in Kenya and Elsewhere to Detect and Analyze Food Types 5

Scraping Social Media Photos Posted in Kenya and Elsewhere t...

引用

5th International Workshop on Multimedia Assisted Dietary Management (MADiMa)

作者： Jalal, Mona Wang, Kaihong Jefferson, Sankara Zheng, Yi Nsoesie, Elaine O. Betke, Margrit Boston Univ Boston MA 02215 USA Lori Syst LTD Nairobi Kenya

ISBN: (纸本)9781450369169

Monitoring population-level changes in diet could be useful for education and for implementing interventions to improve health. Research has shown that data from social media sources can be used for monitoring dietary behavior. We propose a scrape-by-location methodology to create food image datasets from Instagram posts. We used it to collect 3.56 million images over a period of 20 days in March 2019. We also propose a scrape-by-keywords methodology and used it to scrape similar to 30,000 images and their captions of 38 Kenyan food types. We publish two datasets of 104,000 and 8,174 image/caption pairs, respectively. With the first dataset, Kenya104K, we train a Kenyan Food Classifier, called KenyanFC, to distinguish Kenyan food from non-food images posted in Kenya. We used the second dataset, KenyanFood13, to train a classifier KenyanFTR, short for Kenyan Food Type Recognizer, to recognize 13 popular food types in Kenya. The KenyanFTR is a multimodal deep neural network that can identify 13 types of Kenyan foods using both images and their corresponding captions. Experiments show that the average top-1 accuracy of KenyanFC is 99% over 10,400 tested Instagram images and of KenyanFTR is 81% over 8,174 tested data points. Ablation studies show that three of the 13 food types are particularly difficult to categorize based on image content only and that adding analysis of captions to the image analysis yields a classifier that is 9 percent points more accurate than a classifier that relies only on images. Our food trend analysis revealed that cakes and roasted meats were the most popular foods in photographs on Instagram in Kenya in March 2019.

关键词： image datasets food detection data mining social media

来源：评论

学校读者我要写书评

暂无评论

Black gram Plant Leaf Disease (BPLD) dataset for recognition and classification of diseases using computer-vision algorithms

引用

DATA IN BRIEF 2022年 45卷 108725页

作者： Talasila, Srinivas Rawal, Kirti Sethi, Gaurav Sanjay, M. S. S. Reddy, Surya Prakash M. Lovely Profess Univ Phagwara Punjab India VNR Vignana Jyothi Inst Engn & Technol Hyderabad Telangana India Jawaharlal Nehru Krishi Vishwa Vidyalaya Jabalpur India Malla Reddy Univ Hyderabad Telangana India

This article introduces Black gram Plant Leaf Disease (BPLD) dataset, which is scientifically called as Vigna Mungo and is popularly known as Urad in India. It is widely considered to be one of the most significant pulse crops farmed in India. Anthracnose, Leaf Crinkle, Powdery Mildew and Yellow Mosaic diseases shown significant impact on the black gram production and causing financial loss to the farmers. A fusion of image processing and computer vison algorithms are widely used in recent years, for applications in the diagnosis and categorization of diseases that affect plant leaves. To detect and classify plant leaf diseases which degrades the quality of the black gram crop, in early stages, using computer vision algorithms, a Black gram Plant Leaf Disease (BPLD) dataset was created and briefly discussed in this article. The dataset holds a total of 100 0 images belongs to five classes: four diseases and one healthy. The images in the presented dataset were captured under the real cultivation fields at Nagayalanka, Krishna, Andhra Pradesh, using camera and mobile phones. After the image acquisition, the images were categorized and processed with the help of agriculture experts. Researchers who utilize image processing, machine learning and particularly deep learning algorithms for automated diagnosis and classification of black gram plant leaf diseases in early stage to assist farmers could benefit from this dataset. The dataset is publicly and freely available at https://***/10.17632/zfcv9fmrgv.3 (C) 2022 The Author(s). Published by Elsevier Inc.

关键词： Black gram crop (Vigna Mungo) image datasets image classification Machine Learning Deep Learning

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：