检索结果-内蒙古大学图书馆

DePondFi'23 Challenge on Real-Time Pond Environment: Methods and Results

IEEE ACCESS 2024年 12卷 157975-157987页

作者： Sasithradevi, A. Suganya, R. Prakash, P. Mohamed Mansoor Roomi, S. Vijayalakshmi, M. Nathan, Sabari Kasthuri, P. Persiya, J. Brighty Ebenezer, L. Jain, Sparsh Verma, Sshubam Balasubramanian, S. Sai Subramaniam, M. Sai Sriram, T. Pranav Phanindra Sai, M. Raj, Chandan Yadav, Amandeep Payak, Ritik Paul Choudhury, Suman Singh, Rohit Vellore Inst Technol Ctr Adv Data Sci Chennai 600127 India Vellore Inst Technol Sch Comp Sci & Engn Chennai 600127 India Indian Inst Technol Madras Dept Elect Engn Chennai 600044 India Thiagarajar Coll Engn Dept Elect & Commun Engn Madurai 625015 India Vellore Inst Technol Sch Elect Engn Chennai 600127 India Couger Inc Tokyo Japan Maharaja Agrasen Inst Technol Dept Comp Sci & Engn Delhi 110086 India Sri Sathya Sai Inst Higher Learning Dept Math & Comp Sci Anantapur 515134 Andhra Pradesh India IIT Roorkee Dept Elect Engn Roorkee 247667 India SAHAJ AI Bangalore 560034 India

This article summarizes the Detection of Pond Fish Challenge (DePondfi'23 Challenge), held during the National conference on computer vision, Pattern Recognition, image processing and graphics (NCVPRIPG 2023). The main goal of the challenge was to find the most effective methods for detecting pond fish in underwater images, overcoming obstacles such as poor visibility, variations in turbidity, and environmental shifts. Sixty participants registered, with 15 teams submitting results for phase 1. The challenge concluded with four teams earning top honors based on mAP (mean Average Precision) score and time complexity. The mAP scores achieved by toppers are as follows: DETECTRON - 38.93%, DMACS SAI - 36.65%, Pondvision - 31.63%, and Sahajeevis - 29.06%. This article describes the toppers method and discusses the detection results. Our challenge event is in line with Sustainable Development Goal 14, which focuses on the conservation and sustainable utilization of ponds and marine resources for sustainable development.

关键词： Aquaculture underwater challenges detection models mAP computational time.

来源：评论

学校读者我要写书评

暂无评论

Integrated algorithm for different tracking challenges

Integrated algorithm for different tracking challenges

引用

National conference on computer vision, Pattern Recognition, image processing and graphics (NCVPRIPG)

作者： Jay Shah Sai Subrahmanyam G. R. K Deepak Mishra Dept. of Avionics Indian Institute of Space sciences and Technology Trivandrum Kerala India Dept. of Earth & space Science Indian Institute of Space sciences and Technology Trivandrum Kerala India

This paper presents an efficient combination of two well-known tracking algorithms, Tracking-Learning-Detection (TLD) and Compressive Tracking (CT) to devise an algorithm which takes advantages of both and outperforms them on their short-ends by virtue of other. TLD fails in cases including full out-of-plane rotation, fast motion and articulated object tracking. While CT fails in resuming tracking once the object leaves the frame and comes back. We propose a combining algorithm mentioned as Algorithm 1, which robustly handles all the tracking challenges. Different thresholds are set which can be varied to weigh each component as required. The proposed algorithm is tested on different test sequences involving challenging tracking scenarios such as fast motion and their success rates are calculated in Table I. The proposed algorithm works favourably against both algorithms in terms of robustness and success rate.

关键词： Target tracking Detectors Feature extraction Object tracking Robustness Aerospace electronics

来源：评论

学校读者我要写书评

暂无评论

Computational Intelligence in Pattern Recognition 1

引用

丛书名： Advances in Intelligent Systems and Computing

1000年

作者： Asit Kumar Das Janmenjoy Nayak Bighnaraj Naik Soumi Dutta Danilo Pelusi

ISBN: (数字)9789811625435

ISBN: (纸本)9789811625428

This book features high-quality research papers presented at the 3rd International conference on Computational Intelligence in Pattern Recognition (CIPR 2021), held at the Institute of Engineering and Management, Kolkata, West Bengal, India, on 24 – 25 April 2021. It includes practical development experiences in various areas of data analysis and pattern recognition, focusing on soft computing technologies, clustering and classification algorithms, rough set and fuzzy set theory, evolutionary computations, neural science and neural network systems, image processing, combinatorial pattern matching, social network analysis, audio and video data analysis, data mining in dynamic environments, bioinformatics, hybrid computing, big data analytics and deep learning. It also provides innovative solutions to the challenges in these areas and discusses recent developments.

关键词： Computational Intelligence computer Imaging, vision, Pattern Recognition and graphics Artificial Intelligence Mobile and Network Security

来源：评论

学校读者我要写书评

暂无评论

Improving Video Stabilization in the Presence of Motion Blur

Improving Video Stabilization in the Presence of Motion Blur

引用

National conference on computer vision, Pattern Recognition, image processing and graphics (NCVPRIPG)

作者： Manish Okade P.K. Biswas Department of Electronics and Electrical Communication Engineering Indian Institute of Technology Kharagpur India

In this paper we propose the idea of deblurring for efficient feature matching in the context of video stabilization. This is achieved by incorporating a motion deblurring block prior to the feature extraction and matching stage in the stabilization pipeline. The effect of motion blur on feature extraction and matching has not been investigated so far as most of the approaches have looked into deblurring as a post processing step. After preprocessing the blurred frames, the scale invariant feature transform points are used to find correspondence to get an estimate of the camera motion. Smoothing of the motion parameters to retain the desired motion is performed using a gaussian filter. Finally, inverse of the resulting image transform is carried out to obtain a stable video sequence. We compare our method to existing approaches and show how the inserted block improves the video stabilization performance. Interframe Transformation Fidelity (ITF) is used to show the superiority of our proposed approach.

关键词： Feature extraction Cameras Video sequences Motion estimation Indexes Motion compensation Pattern matching

来源：评论

学校读者我要写书评

暂无评论

A novel stereo matching approach for pedestrian re-identification

A novel stereo matching approach for pedestrian re-identific...

引用

National conference on computer vision, Pattern Recognition, image processing and graphics (NCVPRIPG)

作者： Soma Biswas Department of Electrical Engineering Indian Institute of Science Bangalore India

Automatic and reliable identification of pedestrians from multiple camera views is very important for video surveillance and can save a lot of manual effort. The significant variations in viewpoints, poses, illumination and occlusions makes this problem very challenging. Most of the existing approaches addressing this problem handle drastic viewpoint change in a supervised way and thus require labelling new training data for a different pair of camera views. In this paper, we present a novel approach for pedestrian re-identification using stereo matching, which does not require any kind of training. The cost of the stereo matching of two images is used for evaluating the similarity of the images, without performing 3-D reconstruction. We show that this cost is robust to the large pose variations observed in the images captured from multiple cameras. The proposed pedestrian re-identification algorithm is built on top of a dynamic programming stereo matching algorithm. Experimental evaluation on the challenging VIPeR dataset shows the effectiveness of the proposed approach.

关键词： Cameras image color analysis Histograms Heuristic algorithms Lighting Robustness computer aided manufacturing

来源：评论

学校读者我要写书评

暂无评论

Detecting and tracking the position of suspicious objects using vision system

Detecting and tracking the position of suspicious objects us...

引用

conference on Artificial Intelligence for Security and Defence Applications

作者： Muthukrishnan, Ramkumar Kannan, Somasundar Prabhu, Radhakrishna Zhao, Yafan Bhowmick, Parijat Robert Gordon Univ Sch Engn Aberdeen Scotland Indian Inst Technol Elect & Elect Engn Dept Gauhati India

ISBN: (纸本)9781510667136;9781510667143

vision-based object tracking is crucial for both civil and military applications. A range of hazards to cyber safety, vital infrastructure, and public privacy are posed by the rise of drones, or unmanned aerial vehicles (UAV). As a result, identifying suspicious drones/UAV is a serious issue that has attracted attention recently. The key focus of this research is to develop a unique virtual coloured marker based tracking algorithm to recognise and predict the pose of a detected object within the camera field-of-view. After detecting the object, proposed method begins by determining the area of detected object as reference-contour. Following that, a Virtual-Bounding Box (V-BB) is developed over the reference-contour by meeting the minimum area of contour criteria. In order to track and estimate the precise location of the detected object in two-dimensions during observations, a Virtual Dynamic Crossline with a Virtual Static Graph (VDC-VSG) was constructed to follow the motion of V-BB, which is considered as a virtual coloured marker. Additionally, the virtual coloured marker helps to avoid issues linked to ambient lighting and chromatic variation. To some extent, it can function efficiently during obstructions like rapid position fluctuations, low resolution and noises etc. The efficacy of the developed algorithm is evaluated by testing with significant number of aerial sequences, including benchmark footage and the outputs were outstanding, with better results. The suggested method will support future industry of computer vision-based intelligent systems. Potential applications of the proposed method includes object detection and analysis applied to the field of security and defence.

关键词： image processing Cascade classifier Object detection Object tracking Aerial images

来源：评论

学校读者我要写书评

暂无评论

Global vessel symmetry for optic disc detection in retinal images

Global vessel symmetry for optic disc detection in retinal i...

引用

National conference on computer vision, Pattern Recognition, image processing and graphics (NCVPRIPG)

作者： Rashmi Panda N. B. Puhan G. Panda School of Electrical Sciences Indian Institute of Technology Bhubaneswar India

Optic disc (OD) detection is an important step in developing computer aided screening systems suitable for glaucoma analysis. In this paper, we present a new method for automatic optic disc detection in retinal (fundus) images. The method is based upon the distribution of major blood vessels. The blood vessels originate from the OD and their random distribution pattern can be approximately divided into two halves by a global symmetric axis passing through the centroid and near the optic disc. We detect this symmetry axis by using partial Hausdorff distance (PHD) measure. Then, the OD center is detected by applying the brightness property of the optic disc region. The proposed method is evaluated and compared on DRIVE, STARE and HRF databases. The average performance of the proposed method is found as: 97.5% in DRIVE, 97.5% in STARE and 100% in HRF database.

关键词： Blood vessels Retina Optical imaging Databases image color analysis Biomedical optical imaging

来源：评论

学校读者我要写书评

暂无评论

Parallelizing connectivity-based image processing operators in a multi-core environment

Parallelizing connectivity-based image processing operators ...

引用

International conference on Communications and Signal processing

作者： B Ravi Kiran K P Anoop Y Senthil Kumar CVAI Lab Indian Institute of Science Electrical Engineering Bangalore India Texas Instruments India Imaging Group Automotive Computer Vision group Bangalore India

With increasing computational requirements for state-of-art computer vision and graphics algorithms, multi-core processor implementations have picked up momentum. Adapting single core image processing algorithms for multi-core environment is difficult. Different platforms require a hand tailored approach to optimally fit the algorithm into a given multi-core architecture. An automatic compiler which could split and compile the code for any platform is desirable. Such a compiler doesn't exist and this kind of automation is tough to achieve. To aid existing compilers, we propose a simple software architecture framework for multi-core systems that aims to provide a uniform, scalable, and portable solution, solving specifically for connectivity based image processing operators.

关键词： image resolution Multicore processing graphics processing unit Hardware Encoding Labeling

来源：评论

学校读者我要写书评

暂无评论

Iris liveness detection using texture segmentation

Iris liveness detection using texture segmentation

引用

National conference on computer vision, Pattern Recognition, image processing and graphics (NCVPRIPG)

作者： Mohit Kumar N. B. Puhan School of Electrical Sciences Indian Institute of Technology Bhubaneswar India

In order to chisel an iris recognition system and pass in the iris liveness test, an attacker can create semitransparent contact lens spoofing. Such a contact lens is transparent around the iris center and has fake iris texture of another person printed around the outer region. In this paper, such fake iris images are synthetically created which can obviate sleuthing even after using pupillary light reflex technique. Thereafter, new liveness detection is proposed to determine the perceptually invisible boundary between the fake and original iris textures. Response of Gaussian derivative filters with multiple scales and orientations at each pixel location is clustered using K-means to ascertain regions with different textures. To give robustness to the algorithm, it is iterated a certain number of times and a threshold mechanism is imposed to find the correct boundary. The proposed method is shown to achieve high liveness performance by generating a set of 600 fake iris images from the UPOL iris database.

关键词： Iris recognition Lenses Iris image segmentation image resolution Clustering algorithms Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Fast moving-object detection in H.264/AVC compressed domain for video surveillance

Fast moving-object detection in H.264/AVC compressed domain ...

引用

National conference on computer vision, Pattern Recognition, image processing and graphics (NCVPRIPG)

作者： Manu Tom R. Venkatesh Babu Video Analytics Laboratory SERC Indian Institute of Science Bangalore India

This paper discusses a novel fast approach for moving object detection in H.264/AVC compressed domain for video surveillance applications. The proposed algorithm initially segments out edges from regions with motion at macroblock level by utilizing the gradient of quantization parameter over 2D-image space. A spatial median filtering of the segmented edges followed by weighted temporal accumulation accounts for whole object segmentation. To attain sub-macroblock (4×4) level precision, the size of macroblocks (in bits) is interpolated using a two tap filter. Partial decoding rules out the complexity involved in full decoding and gives fast foreground segmentation results. Compared to other compressed domain techniques, the proposed approach allows the video streams to be encoded with different quantization parameters across macroblocks thereby increasing flexibility in bit rate adjustment.

关键词： Streaming media Motion segmentation conferences Video surveillance Quantization (signal) image coding

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：