气泡识别率达到95%以上，能够满足工业检测的实际需要。
The bubble recognition rate is higher than 95%, so it can meet the practical need of industry detection.
介绍一种利用灰度图像识别玻璃瓶气泡的方法。
Introduce a method of identify bubble in glassware based gradation graph.
由于这一技术受到气泡流速的影响，容易造成流型识别的混淆。
However this method easily makes a confusion in flow regime identification owing to the influence of gas speed.
对于清晰的高分辨率气泡图像，本文提出了一种新的气泡边界识别方法，能够自动地得到气泡的闭合边界。
The paper brings forward new bubble boundary identification method for High resolution bubble image, which can automatically receive bubbles closed boundary.
在实验的基础上提出了一种用激光后向散射信号的幅度谱进行探测和识别气泡的新途径。
After the spectral analysis of back scattering we can identify the bubbles by the differences between bubbles and water.
本文基于一种全新的测试方法“气泡幅频当量法”，研制了一种炉前快速识别系统。
In this paper, a testing system for fast predicting graphic shape in front of furnace, based on a new method named "amplitude and frequency equivalent of bubbles", will be introduced in detail.
根据气泡和水对光的反射特性不同，提出采用阈值法识别图像中的气泡和水，从而为进一步计算掺气浓度打下坚实的基础。
According to different optics-dispersion character, threshold segmentation method is proposed to identify gas from the gas-water flow, which make steadfast base for further calculate void fraction.
本系统采用最优阈值变换提取目标物体轮廓，完成了一个能够实现对气泡的偏移进行检测和对图像的状态进行模式识别的图像处理和识别系统。
The calibration system utilizes the combination of advantages of optimal thresholding and contour tracking to analyze the image of air bubble and recognize its features.
本系统采用最优阈值变换提取目标物体轮廓，完成了一个能够实现对气泡的偏移进行检测和对图像的状态进行模式识别的图像处理和识别系统。
The calibration system utilizes the combination of advantages of optimal thresholding and contour tracking to analyze the image of air bubble and recognize its features.
应用推荐