智能感知技术是智能制造落地应用及生产过程控制、预报及工艺优化的基础，近年来随着机器视觉及人工智能技术的快速发展，在钢铁领域已有大量基于机器视觉的典型应用，比如温度检测、表面缺陷检测、尺寸检测、废钢识别、辅助定位、物流号识别等，对提高钢铁行业质量检测智能化、推动智能制造生根落地、提高生产效率及产品质量起到了良好的推动作用。针对目前国际前沿的机器视觉智能感知技术在钢铁行业的广泛应用情况，本书将首次成体系介绍基于机器视觉的智能感知技术在钢铁行业应用的多学科交叉理论体系和部分研究成果。

1 Temperature Detection Methods Based on Machine Vision
1.1 Colorimetric thermometric model
1.1.1 Basic assumptions
1.1.2 The relationship of image grayscale and radiant temperature
1.1.3 Colorimetric temperature measurement model
1.2 Blackbody furnace calibration process
1.3 Image noise filtering method
1.3.1 Adaptive smooth filter denoising method
1.3.2 Adaptive median filter denoising method
1.3.3 Geometric mean filtering denoising method
1.3.4 Over-limit neighborhood filtering denoising method
1.3.5 Bilateral filter denoising method
1.3.6 Wavelet filter denoising method
1.4 Image edge detection method
1.4.1 Roberts operator
1.4.2 Sobel operator
1.4.3 Prewitt operator
1.4.4 Log operator
1.4.5 Canny operator
1.4.6 Morphological edge detection method
2 Temperature Detection in the Raceway Zone of Blast Furnace
2.1 Introduction
2.2 Theory and experiments
2.3 Methods to improve the temperature detection accuracy
2.3.1 Blackbody calibration for digital imaging system
2.3.2 Denoising for tuyere images
2.3.3 Edge detection for tuyere images
2.4 Combustion behavior and temperature in tuyere zone of BF
2.5 PCI process on the temperature variation in raceway zone of BF
2.5.1 Flame image temperature calculation
2.5.2 PCI rates
2.5.3 PCI cease processes
2.5.4 None PCI process
References
3 Uniformity and Activity of Blast Furnace Hearth
3.1 Introduction
3.2 Definitions of uniformity index and activity index
3.3 Experiments and temperature calculation
3.4 Uniformity index and activity index
3.4.1 Local uniformity index and local activity index
3.4.2 Ul and AI
References
4 Tuyere Coke Size and Temperature Distribution of BF
4.1 Introduction
4.2 Theory
4.2.1 Tuyere coke size detection
4.2.2 Temperature distribution calculation
4.3 Experiments
4.4 Results of tuyere coke size and temperature distribution
4.4.1 Calculation of tuyere coke size and temperature distribution of raceway zone
4.4.2 Results analyze
4.4.3 Calculation of coke belt length (CBL) in raceway zone of BF
References
5 Surface Temperature of Rail Steel Plates
5.1 Introduction
5.2 Emissivity calculation model
5.3 Theory
5.4 Experiments
5.4.1 Digital imaging system
5.4.2 Calibration for digital imaging system
5.4.3 Surface temperature variation during the rolling process
5.5 Temperature calculation results
5.5.1 Noise filtering results
5.5.2 Temperature detection results during rolling passes
References
6 Surface Defect Recognition of Aluminum Strips
6.1 Introduction
6.2 Aluminum strips images
6.3 NSST-KLPP model
6.4 Experiment and discussions
6.4.1 Experimental procedure
6.4.2 Experiment results
6.4.3 Comparison with common methods
6.4.4 Comparison of NSST,contourlet and DST
References
7 Surface Inspection of Continuous Casting Slabs
7.1 Introduction
7.2 Surface defects
7.3 DNST-GLCM-KSR model
7.4 Defect recognition algorithm
7.5 Defect recognition results
References
8 Surface Defect Recognition of Metals
8.1 Introduction
8.2 Construction of ICT
8.3 Defect recognition algorithms
8.4 Defect recognition results
References
9 Edge Detection of Retinal OCT Image
9.1 Introduction
9.2 Complex shearlet transform
9.3 Construction of detection formula
9.4 Numerical and visual experiments
9.5 Retinal OCT experiment
References
10 Detection of Character and Surface Defect in Steel Rail
10.1 Introduction
10.2 Experimental platform and experimental method
10.3 Methods
10.3.1 Photometric stereoscopic 3D reconstruction method
10.3.2 Character recognition method based on deep learning
10.4 Detection results
10.4.1 Defect detection result