航空发动机损伤图像的二分类到多分类递进式检测网络

航空发动机损伤是影响飞行安全的重要因素。当前基于计算机视觉的发动机孔探图像损伤检测存在两个主要问题：一是孔探图像背景复杂,使得模型对损伤的检测精度较低;二是孔探图像数据来源受限,导致模型可检测类别较少。为解决这两个问题,提出了基于Mask R-CNN的二分类到多分类递进式航空发动机损伤图像检测网络。通过在Mask R-CNN中增加二分类检测分支,首先对图像中的损伤进行二分类检测并对定位坐标进行回归优化;其次使用原始检测分支递进地进行多分类检测,以进一步回归优化损伤的检测结果并确定损伤类型;最后根据多分类检测的结果,通过Mask分支对对损伤进行实例分割。为了增加模型检测类别及验证方法的有效性,构建了包含八种损伤类型,共1 315张孔探图像的数据集。在该集合上进行的训练和测试结果表明,多分类检测的平均精度（AP）和AP75与Mask R-CNN相比分别提高3.34%、9.71%,可见所提方法能够有效提高对孔探图像中的损伤的多分类检测精度。     

自适应参数分析结果的航空发动机孔探图像损伤检测

目的 航空发动机孔探图像的损伤检测关系到航空发动机是否要非例行更换,直接影响飞机的飞行安全和利用率。现有的孔探图像损伤检测方法直接使用目标检测方法训练一个多类别损伤检测器,使用相同的参数在不同位置检测损伤。由于没有考虑同类型损伤在发动机不同区域发生概率的不同,导致现有方法的检测准确率较低。为了提高损伤检测的准确率,提出了一种自适应参数的航空发动机孔探图像损伤检测方法。方法 通过识别孔探图像所属的发动机区域,针对不同区域孔探图像设置不同的参数用于检测发动机损伤。同时为了避免单检测器上不同类型损伤之间相互干扰,采用独立检测器检测单一类型的损伤,并对误检率高的损伤进行真假识别。通过合并检测到的不同类型的损伤,得到最终的损伤检测结果。此外,为了改进水平的矩形检测框,使用分割结果产生旋转的检测框,有效地减少了框中的背景区域。结果 在13个航空发动机区域的2 654幅孔探图像上针对烧蚀、裂缝、材料丢失、涂层脱落、刻痕和凹坑等6种典型的发动机损伤进行检测实验。提出的损伤检测方法在准确率和召回率两方面分别达到了90.4%和90.7%,相较于目标检测方法YOLOv5 （you only look once version 5）的准确率和召回率高24.8%和25.1%。实验结果表明,本文方法在航空发动机损伤检测方面优于其他对比方法。结论 本文所提出的自适应参数的航空发动机损伤检测模型通过识别发动机图像所属的部位,针对同种类型的损伤检测器设定不同的参数,有效地提高了检测器的检测性能。同时,针对容易误检的裂缝、刻痕和凹坑增加了真假损伤判别器,有效地减少了误检的情况。     

基于DenseNet和ResNet融合的发动机孔探图像分类研究

孔探是检测发动机内部损伤最重要的手段之一.为了解决发动机孔探检查中孔探人员主要依靠经验对损伤进行界定的问题,研究了基于DenseNet和ResNet融合的新型单通道网络结构,实现对发动机部件的分类,为后期孔探缺陷自动识别建立基础.通过对某大修厂孔探数据和自建数据进行处理,完成了孔探图像分类数据集的构建;训练新型的49层网络模型,在自建数据集测试集上测试的准确率和平均召回率分别为96.0％ 和95.9％,有较好的泛化能力,可以有效的对发动机部件进行分类.     

航空发动机温度测量电路设计

为了提高航空发动机温度测量的精度和可靠性,给出了一种航空发动机温度测量电路的设计方法。该方法首先对航空发动机温度测量技术要求进行分析;进而设计了航空发动机低温、高温温度测量系统方案;最后采用模块化设计方法给出了关键的信号调理电路。该电路设计方法能满足航空发动机对不同温度环境测试的需求,有利于提高航空发动机工作温度测量系统的性能,具有一定的实际参考价值。     

基于机器视觉的航空发动机转子裂纹故障识别系统

在大型设备早期故障诊断中,利用数字图像处理可以很好地对设备状态做出测试和分析,以航空发动机转子表面早期裂纹故障为例,首先利用CCD图像采集将被摄取目标转换成图像信号,其次通过像素分布和亮度、颜色分析,将这些信息转变成数字信号,再次利用现代数字信号处理技术对这些信号进行运算来提取目标特征,最后对设备的故障状态做出判断;以图像获取、预处理、图像分割、裂纹特征提取和故障识别为算法流程,设计了一套转子系统早期裂纹自动检测识别系统,实验分析表明,该系统在航空发动机转子早期表面裂纹故障检测识别中具有精度高、速度快的优点。     

8通道静态应变信号调理模块研究

在航空发动机的测试中,需要测量发动机的各种性能参数,如温度、压力、转速、流量、振动等,其中压力的测试是可以通过测量发动机叶片产生的应变来完成,通常应变测试系统包括应变片、信号调理器、数据采集模块和软件处理几部分;北京长城测控公司结合近几年开发航空发动机测试系统的经验,基于VXI总线测试技术开发了信号调理模块与数据采集模块高度集成的航空发动机测试系统,同时针对测试中存在的多种物理量开发了多种信号调理模块;文章主要针对静态应变信号调理模块的使用方法及在数据采集系统中的应用进行了介绍,并对关键电路的设计方法进行了详细说明。     

航空发动机转子扭转振动测试仪的研究

讨论了航空发动机扭转振动测量的特点及其对测量系统的要求,提出一种基于激光多普勒原理的新型航空发动机转子扭转振动测量方法,并分析了影响该方法测量精度的因素、测量范围,实验结果表明该测量方法能满足航空发动机转子扭转振动测量的要求。     

一种基于ASAP3协议的自动测试软件开发

介绍了用于发动机自测试工作的测量与标定功能接口协议ASAP3,应用该协议开发了一款自动测试软件,与发动机台架、标定工具以及相关执行机构组成了一套自动测试系统。自动测试软件通过指令程序控制发动机工作状态,同时通过TCP/IP网络协议访问测量标定工具,读取ECU内部实时数据并存储,实现对发动机进行在线自动测量和标定,解决了通用测量标定工具不能同步调整发动机状态并自动测量的问题。最后应用该自动测试系统对GPF材料进行了老化验证试验。     

基于图像融合的航空发动机温度场检测技术研究

通过对航空发动机温度场常用检测方法进行分析,论文首次提出了一种基于多传感器图像融合航空发动机温度场检测方法。实验证明,该方法大大地提高了温度场的检测精度,也提高了系统的鲁棒性。     

基于ICT切片图像的航空发动机涡轮叶片壁厚检测

该文针对航空发动机中的涡轮叶片的检测问题,研究了基于ICT切片图像的叶片壁厚测量方法。该方法主要在关键点选取上进行了较好的筛选,并对结果进行了拟合。实验结果表明,采用该测量方法可以保证检测误差小,测量结果直观等特点。从而较好地满足了航空发动机中对叶片壁厚检测的要求。     

橡胶密封件裂缝缺陷定位方法研究

为了实现在橡胶密封件表面缺陷的自动检测中快速有效地定位裂缝缺陷,根据密封件裂缝缺陷图像的特点,应用自适应阈值分割算法,结合改进的二值形态学并行细化算法,提出了一种橡胶密封件裂缝缺陷的定位方法。该方法可从复杂的密封件图像中分割出裂缝区域,并完成对裂缝缺陷的定位。实验表明,该方法获得的定位结果与实际裂缝位置相符,对于橡胶密封件表面缺陷检测与质量提高具有重要意义。     

Automatic control of casting speed in ingot casting

A microwave casting speed meter and the automatic casting speed control system that uses it have been developed for the bottom pouring process of steel ingots. The microwave meter is based on the Doppler radar technique, and a circuit of period measurement and reciprocal calculation has been developed to make its response time faster. Also it has a function to measure cast height. The automatic system can be regarded as a sampled-data system with an adaptive control function. Outlines of the microwave meter, the control circuit and the method of adaptive control are presented. The relationship of the crack appearance to the casting speed is studied, and programmed patterns of the casting speed have been prepared to reduce cracking. The programmed ‘pattern casting’ has been achieved by the automatic control system, and the surface quality of the ingot has been improved.     

基于LabVIEW的钢杆裂纹定量识别技术的研究

介绍基于LabVIEW的钢材表面裂纹定量识别技术，设计了基于LabVIEW的钢杆裂纹定量检测系统，并时钢杆表面横向裂纹缺陷进行模拟实验。初步建立数学模型，通过评判得到反映裂纹状况的定量检测结果。从而验证了该方法的可行性和有效性。并得出相关的结论。     

Inspection equipment study for subway tunnel defects by grey-scale image processing

In recent years, much attention has been paid to Machine Vision-Based (MVB) technology for tunnel main defect (leakage and crack) inspection as an innovative technology. Based on the principle of MVB technology, various researchers have developed tunnel inspection equipment, but most of them need either a trailer or an external power supply, which cannot meet the demand of subway tunnel inspection in China. The limited inspection time, high demand for precision, rigid requirements of operational management and high cost of the equipment restrict the application of this method in China. MTI-100 (Moving Tunnel Inspection) was developed under these circumstances. To capture stable, high-quality images of the lining surface as the raw data of inspection, an image capture system is well designed based on CCD (Charge-coupled Device) camera scanning. Additionally, equipment optimization design of the mechanism and electricity requirements for the inspection accuracy of subway tunnel inspection is investigated. The maximal size and weight of equipment elements determined the convenience of inspection, which is primarily conditioned by these designs. The effects of lighting and vibration have been considered. A method to calculate the image shift caused by vibration is proposed. The software network is another core component of the equipment, which connects the image acquisition, image storage and defect recognition. The famous Otsu method is used for leakage recognition. A new algorithm based on the features of the local image grid is developed to recognize cracks. A comparative study shows its high accuracy for crack recognition. Finally, a simulative tunnel test and field inspection are undertaken to verify the performance of the non-destructive subway tunnel inspection equipment. Through these tests, the accuracy, stability, repeatability, labor intensity and efficiency of the equipment have been verified. A real project test certified that the developed MTI-100 is quite suitable for practical tunnel inspection.     

一种基于改进渗流模型的混凝土表面裂缝快速检测算法

由于混凝土表面不平整、光照不均、裂缝背景较为复杂等因素的干扰,传统的基于图像处理的裂缝检测方法对裂缝检测的效果不佳,尤其是不清晰和比较细小的裂缝。基于渗流模型的裂缝检测方法充分考虑了裂缝亮度低、形状较为细长的特点,对裂缝的检测效果很好,尤其是图像中的细小裂缝,但是该方法需要大量的处理时间。为了解决上述问题,提出了一种加速渗流处理的算法,该算法通过暗点预提取来减少渗流时需要处理的像素点个数,以此来减少渗流处理时间。实验结果表明,所提算法能明显加快渗流处理的速度,并且精确率基本保持不变。     

An innovative methodology for detection and quantification of cracks through incorporation of depth perception

Visual inspection of structures is a highly qualitative method in which inspectors visually assess a structure’s condition. If a region is inaccessible, binoculars must be used to detect and characterize defects. Although several Non-Destructive Testing methods have been proposed for inspection purposes, they are nonadaptive and cannot quantify crack thickness reliably. In this paper, a contact-less remote-sensing crack detection and quantification methodology based on 3D scene reconstruction (computer vision), image processing, and pattern recognition concepts is introduced. The proposed approach utilizes depth perception to detect cracks and quantify their thickness, thereby giving a robotic inspection system the ability to analyze images captured from any distance and using any focal length or resolution. This unique adaptive feature is especially useful for incorporating mobile systems, such as unmanned aerial vehicles, into structural inspection methods since it would allow inaccessible regions to be properly inspected for cracks. Guidelines are presented for optimizing the acquisition and processing of images, thereby enhancing the quality and reliability of the damage detection approach and allowing the capture of even the slightest cracks (e.g., detection of 0.1 mm cracks from a distance of 20 m), which are routinely encountered in realistic field applications where the camera-object distance and image contrast are not controllable.     

基于LabVIEW的透镜多参数自动检测系统开发

质量检测是光学透镜生产的重要环节,目前大部分透镜生产企业还停留在半自动检测阶段,质检的效率和精度很大程度上取决于工人的熟练程度;且不同参数的检测需要人工搬运镜片到不同的质检工位上,质检过程中的反复搬运容易造成镜片表面污染和损伤;为此提出一种基于LabVIEW的透镜多参数自动检测系统的设计方法,在一个工位上同时完成多个质量参数的测量,并对透镜进行自动搬运和分选,检测结束后自动生成质检报表;最后使用某型号凸型圆形透镜对开发的检测系统进行了实验验证,实验结果表明,检测系统具有测量精度高、可扩展能力强、界面直观等优点。     

Fuzzy set based crack diagnosis system for reinforced concrete structures

This paper presents a computer assisted crack diagnosis system for reinforced concrete structures which aids the non-expert to diagnose the cause of cracks at the level of an expert in the general inspection of structures. The system presented adapts fuzzy set theory to reflect fuzzy conditions, both for crack symptoms and characteristics which are difficult to treat using crisp sets. The inputs to the system are mostly linguistic variables concerning the crack symptoms and some numeric data about concrete and environmental conditions. Using these input data and based on built-in rules, the proposed system executes fuzzy inference to evaluate the crack causes under consideration. The built-in rules were constructed by extracting expert knowledge, primarily from technical books about concrete and concrete cracks. We implemented the proposed system in a computer program with a graphic user interface for actual utilization in practical business fields. When applied to cracks actually diagnosed by experts, the proposed system provided results similar to those obtained by experts, and we expect that this system can be used as an effective crack diagnosis tool for both experts and non-experts in the regular inspection of RC structures.     

基于计算机视觉的海产品抗生素残留检验方法

针对海产品抗生素残留传统检验方法存在劳动强度大、重复性高的问题,研究并建立了基于计算机视觉的抗生素残留自动检验系统,该系统包含抑菌圈图像的自适应降噪和背景校正、多分辨率的二次抑菌圈检测拟合、以及符合国家标准的抗生素残留计算方法.以海产品中四环素族抗生素残留检测进行的试验表明,本文研究的方法具有速度快、精度高、操作简便等优点,开发的系统能够满足部分抗生素残留的检测需要.理论分析和试验证明了该系统的优点和有效性.     

轮毂表面自动探伤系统设计

综合运用图像处理中的边缘检测、形态学处理、频域滤波、最优阈值分割等基本方法,在VC 6.0环境下,实现了轮毂表面在线自动探伤系统。该系统能有效地抑制背景光照不均和刀纹、水渍、油渍等干扰的不利影响,较好地实现了对裂纹特征的检测。实验表明,系统所采用的方法具有运算速度快、误判漏判率低、满足在线检测需求的特点。