Laser & Optoelectronics Progress, Volume. 62, Issue 10, 1000006(2025)
Development and Application of Rocket-Borne Image-Measurement Technology in Aerospace
Figures & Tables(15)
![View of outside of rocket and plume upon takeoff, using Imperx cameras[20]](/Images/icon/loading.gif){kind=icon}
Fig. 2. View of outside of rocket and plume upon takeoff, using Imperx cameras[20]
Fig. 3. Optical equipment carried on rocket and captured images[22]. (a) Optical equipment carried on rocket; (b) booster image; (c) booster servo image; (d) second-stage engine image
Table 1. Comparison of traditional image denoising methods
View tableTable 1. Comparison of traditional image denoising methods
Type Name Implementation mode Advantage Shortcoming Application Spatial domain filter Mean filter［ 43 ］Replace each pixel value with average of the pixels in its neighborhood Simple and easy to implement， fast calculation speed， suitable for preliminary denoising Easily result in blurred images， lost edges and details Remove Gaussian noise or other uniformly distributed noise Median filter［ 44 ］Replace each pixel value with the median of the pixel values in its neighborhood Particularly effective， it can effectively remove noise while preserving edge information Denoising effect for Gaussian noise is not as good as mean filter， and computational complexity is high Salt and pepper noise removal， especially for images that contain abrupt noise Bilateral filter［ 45 ］New value of each pixel is weighted average of pixels in its neighborhood， and the weight is determined by the spatial distance and difference in pixel values Remove noise while preserving image edges Computational complexity is high， processing speed is slow， has a great influence on parameter selection for effect Used for image smoothing， noise reduction and edge protection Total variation denoising［ 46 ］Noise is removed by minimizing the total variation of the image （i.e. absolute sum of image gradient） Effectively remove noise while preserving the edge structure of image High computational complexity， parameter selection， such as the weight of total variation has a great influence on results Scenes used for de-noising and where image structure needs to be preserved Frequency domain filter Gaussian low-pass filter［ 47 ］Image is transformed from the spatial domain to frequency domain using Gaussian function to weight frequency components， and then transformed back to the spatial domain No ringing phenomenon， can smoothly remove high-frequency noise Edges and details can be blurred and processed relatively slowly For edges and details can be blurred， processing speed is relatively slow Wiener filter［ 48 ］Based on statistical characteristics of signal and noise， weights of frequency components are adjusted by an adaptive method Can adaptively adjust filter， effect is better Some a priori knowledge of statistical properties of noise and signals is required Suitable for situations where noise model is known or can be estimated
Tools
Get Citation
Copy Citation Text
Yue Wang, Yanhong Jing, Haifeng Zhang, Zeyu He, Fengying Yue, Sen Dong, Weining Chen. Development and Application of Rocket-Borne Image-Measurement Technology in Aerospace[J]. Laser & Optoelectronics Progress, 2025, 62(10): 1000006
Paper Information
Category: Reviews
Received: Sep. 24, 2024
Accepted: Oct. 28, 2024
Published Online: May. 6, 2025
The Author Email: Haifeng Zhang (Zhanghf@opt.ac.cn)
CSTR:32186.14.LOP242017
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20