Passive Millimeter Wave Image Denoising Based on Adaptive Manifolds

Abstract

Abstract-Since the characters of poor inherent resolution and low signal-to-noise limit the application of the passive millimeter wave (PMMW) image, it is particularly important to improve the resolution and denoise the PMMW image. In this paper, the adaptive manifolds filtering algorithm based on non-local means (AM-NLM) is illustrated in detail. And an improved version of AM-NLM filtering algorithm is proposed for processing the PMMW image. The proposed algorithm firstly applies the AM-NLM filtering to obtain the basic denoised PMMW image. Then the image enhancement based on Laplacian of Gaussian operator is performed to enhance the edge of the target in PMMW image. Finally, the hard-threshold filtering with different thresholds is adopted to filter each dimension to achieve the final filtering response. Experimental results have shown that the proposed PMMW filtering algorithm has better and more satisfactory performance compared to AM-NLM, both in subjective visual effect and objective image quality metric. Additionally, our proposed algorithm is also available for real PMMW images.

Multimodal Medical Image Fusion by Adaptive Manifold Filter

“…The adaptive manifold filter is the first high-dimensional filter for performing high-dimensional filtering of images and videos in real time [ 13 ]. The adaptive manifold filter is quite flexible and capable of producing responses that approximate to either standard Gaussian filters or non-local-means filters.…”

A Novel Passive Millimeter Wave Image Noise Suppression Method Based on Pixel Non-local Self-similarity

“…At the same time, PMMW imaging systems do not actively emit electromagnetic waves, making them highly covert and difficult to detect. Therefore, they are widely used in security imaging, military reconnaissance, and battlefield environment sensing [1][2][3][4][5]. However, practical PMMW images often suffer from severe noise, low resolution, and blurring compared to IR and optical images.…”

“…However, practical PMMW images often suffer from severe noise, low resolution, and blurring compared to IR and optical images. These issues arise from factors such as antenna aperture size, diffraction effects, and partial coherence between targets, making it challenging to identify targets effectively from the scene [2]. Therefore, noise suppression is one of the critical challenges for PMMW imaging systems.…”

“…Currently, there are two main research focuses on PMMW image denoising. One approach involves modifying the inversion algorithm for different imaging mechanisms [6][7][8][9][10], while the other aims to improve image quality based on image features [2,[11][12][13]. For instance, an adaptive reconstruction method has been proposed for the total-power radiometer imaging mechanism [8].…”