振动环境下运动目标图像精确检测仿真  

Simulation of Moving Target Image Detection in Vibration Environment

作　　者：邹国华[1] ZOU Guo - hua(East China University of Technology, School of software, Nanehang Jiangxi 330013, China)

机构地区：[1]东华理工大学软件学院,江西南昌330013

出　　处：《计算机仿真》2017年第6期453-457,共5页Computer Simulation

摘　　要：对振动环境下运动目标图像精确检测,可以对运动目标图像中的信息进行有效识别。进行运动目标图像检测时,在振动环境下图像呈现正常和模糊区域,需要将正常区域与模糊区域进行有效分割,有针对性的对模糊区域进行检测,但是传统方法根据点能够判断运动模糊方向,接着采用聚类方法,排除错误点,完成振图像检测,但是不能对图像模糊区域进行有效分割,降低了振动环境下运动目标图像检测效率。提出一种振动环境下运动目标图像精确检测方法,利用傅里叶变换法将振动模糊图像转换至频域,获得具有黑白图像的长条纹,利用变换后的二值化图像运用模糊方向及模糊长度估计算法精确估计运动模糊参数;运用改进的奇异值分解法结合图像抠图技术的对模糊图像正常区域和模糊区域进行分割,观察清晰图像和模糊图像在图像区域分布范围内的差别,标识模糊区域,根据抠图算法再对模糊区域进行精确检测。仿真结果表明,对模糊参数进行精确估计后,利用奇异值分解法及抠图技术算法实现了振动环境下运动目标图像精确检测。This article proposes a precise detection method for moving target image under vibration environment. Firstly, the paper used Fourier transformation to covert vibration fuzzy image into frequency domain and obtained long stripe having monochrome picture. Then fuzzy direction and fuzzy length estimation algorithm was used to estimate the moving fuzzy parameter precisely using the converted binary image. Moreover, the modified singular value decomposition method was used to segment the normal and fuzzy area of fuzzy image integrated with image matting technology, and difference between distinct image and fuzzy image in distribution range of image area was observed. In addition, the research identified fuzzy area. Finally, the fuzzy area precis＇ely was detected according to image matting algo-rithm. Simulation results show that precise detection on moving target image under vibration environment can be achieved using singular value decomposition method and image matting technology after estimating fuzzy parameter precisely.

关 键 词：噪声环境 模糊图像 图像区域分割 识别 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]