paper-with-me

홈 › Papers

Flame quality monitoring of flare stack based on deep visual features

2024-10-18 · Xing Mu

Flare stacks play an important role in the treatment of waste gas and waste materials in petroleum fossil energy plants. Monitoring the efficiency of flame combustion is of great significance for environmental protection. The traditional method of monitoring with sensors is not only expensive, but also easily damaged in harsh combustion environments. In this paper, we propose to monitor the quality of flames using only visual features, including the area ratio of flame to smoke, RGB information of flames, angle of flames and other features. Comprehensive use of image segmentation, target detection, target tracking, principal component analysis, GPT-4 and other methods or tools to complete this task. In the end, real-time monitoring of the picture can be achieved, and when the combustion efficiency is low, measures such as adjusting the ratio of air and waste can be taken in time. As far as we know, the method of this paper is relatively innovative and has industrial production value.

📄 PDF Abstract BibTeX arXiv:2410.19823

Code (0)

등록된 구현이 없습니다.

Tasks

Image SegmentationSemantic Segmentation

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Residual Connection 설명 없음
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…

Similar Papers 제목 키워드 기반

Predicting Solar Flares Using CNN and LSTM on Two Solar Cycles of Active Region Data

2022-04-07 · Zeyu Sun, Monica G. Bobra, Xiantong Wang, Yu Wang 외

We consider the flare prediction problem that distinguishes flare-imminent active regions that produce an M- or X-class flare in the future 24 hours, from quiet active regions that do not produce any flare within $\pm 24…

Attribute

FlameVQA: A Physically-Grounded UAV Wildfire VQA Benchmark with Radiometric Thermal Supervision

2026-06-25 · Mobin Habibpour, John Spodnik, Niloufar Alipour Talemi, Fatemeh Afghah arxiv

Wildfire monitoring from UAVs requires reliable reasoning over complex aerial scenes, where smoke, scale variation, and occlusions often limit RGB-only interpretation. We introduce FlameVQA, a multiple-choice visual ques…

Visual Question Answering

Flame-state monitoring based on very low number of visible or infrared images via few-shot learning

2022-10-14 · Ruiyuan Kang, Panos Liatsis, Dimitrios C. Kyritsis

The current success of machine learning on image-based combustion monitoring is based on massive data, which is costly even impossible for industrial applications. To address this conflict, we introduce few-shot learning…

Few-Shot Learning

SLCFormer: Spectral-Local Context Transformer with Physics-Grounded Flare Synthesis for Nighttime Flare Removal

2025-12-17 · Xiyu Zhu, Wei Wang, Xin Yuan, Xiao Wang arxiv

Lens flare is a common nighttime artifact caused by strong light sources scattering within camera lenses, leading to hazy streaks, halos, and glare that degrade visual quality. However, existing methods usually fail to e…

Flare Removal

FLAME Diffuser: Wildfire Image Synthesis using Mask Guided Diffusion

2024-03-06 · Hao Wang, Sayed Pedram Haeri Boroujeni, Xiwen Chen, Ashish Bastola 외

Wildfires are a significant threat to ecosystems and human infrastructure, leading to widespread destruction and environmental degradation. Recent advancements in deep learning and generative models have enabled new meth…

Fire DetectionImage Generationobject-detectionObject Detection+1