基于分数阶的多向微分算子的高炉料面轮廓自适应检测

高炉料面图像含有丰富的高炉炉况信息，高炉料面轮廓能直接反映料面的凹凸起伏、煤气流分布以及炉况等信息，但高炉料面图像具有对比度低、细节不明显和有强亮斑等特点，使得高炉料面轮廓检测十分困难.本文提出一种新型的高炉料面轮廓检测方法:1）对高炉料面图像进行预处理，增强图像动态范围和图像边缘信息；2）采用分数阶的多向微分算子提取一组料面轮廓可行域；3）用自适应方法确定最佳分数阶阶次，获得可行域中最优的料面轮廓曲线；4）用改进的Canny算子对其进行修正和补偿，得到连续准确的料面轮廓曲线.理论研究和实验结果表明，该方法可准确获取平滑的高炉料面轮廓，对高炉操作人员及时有效调控布料具有很好的参考价值.

Adaptive Detection of Blast Furnace Surface Contour with Fractional Multi-directional Differential Operator

Blast furnace image contains abundant furnace condition information and blast furnace surface contour can directly reflect the bump ups and downs of burden surface, the gas distribution and other information, but the blast furnace burden surface image has the features of low contrast, inconspicuous details and strong bright spots, which make it difficult to detect the blast furnace surface contour. In this connection, a new blast furnace surface contour detection method is proposed. Firstly, the image is preprocessed to enhance its dynamic range and edge information; secondly, multi-directional differential operators based on fractional are deduced to extract a set of blast furnace burden surface contours of feasible region; then, the optimum fractional order is determined by adaptive method to obtain the optimal surface contour curve in the feasible region; lastly, an improved Canny operator is proposed to correct and compensate the optimal surface contour curve. Theoretical research and experimental results show that the new method can accurately obtain a smooth blast furnace burden surface contour, which has great guiding significance for blast furnace foreman to control charging in time and effectively.