An Improved Superpixel-based Fuzzy C-Means Method for Complex Picture Segmentation Tasks

Abstract

Fuzzy c-means(FCM) has attracted wide attentions on picture segmentation as its fuzzy attribute matches the histogram distribution of a picture. However, the fuzzy c-means for the segmentation of a picture with massy noises is barely investigated. In this paper, an improved superpixel-based fuzzy c-means is proposed to segment a massy noise corrupted picture into more than two classes. Firstly, bilateral filtering is used to reduce the compact of noises and makes the picture smoother. Secondly an adaptive method is proposed to fuse the features of the original picture with filtered features. Thirdly simple linearly iterative clustering(SLIC) is used to detect the edge of the picture to avoid over-segmentation. Finally, the histogram-based fuzzy cmeans is used to get the segmentation result. In the experiments, the results show the proposed method achieves a 0.004 ∼ 0.014 higher mPA and 0.004 ∼ 0.06 higher mIoU than other seven algorithms. Besides the segmentation results also show that the over-segmentation is reduced.

A Novel Steganography Technique for Digital Images Using the Least Significant Bit Substitution Method

“…In perspective 4, the proposed method is critically analyzed on three different types of images such as Gray-scale, texture, and aerial images. However, the experimental results prove the improvement and edge of the proposed method which is PSNR values of the Proposed method Image [1] 75.876 Image [2] 79.453 Image [3] 87.980 Image [4] 83.876 Image [5] 89.890 16000 16 KB's Image [6] 90.676 Image [7] 85.897 Image [8] 86.787 Image [9] 77.787 Image [10] 89.787 Image [11] 89.789 Image [12] 90.676 verified through high PSNR values. In addition, the results also prove the strengthening of the proposed method.…”

“…However, for strength and better flawlessness against image preparing activities (trimming, scaling, and Noise assaults) and factual assaults, various research methods have been developed in spatial domain such as: (i) Modified LSB Matching technique [1]; (ii) Enhanced Modified Signed Digit (EMSD) [27]; (iii) Development of the Least Significant Bit (DLSB) [3]; (iv) Pixel Indicator Technique (PIT) [19]; (v) LSB-S and Neighboring Pixel Value Differencing [21]; (vi) Inverted LSB using Adaptive Pattern Technique [2]; and (vii) Pixel Value Differencing (PVD) technique [30] etc. These methods are for the most part utilized and worked well in image steganography and applications because of their better strength against factual steganalysis [12], [11]. However, these techniques have unacceptable amount of embedding message, that consequently result in stego of bad quality images, and computation complexity when contrasted with other spatial domain techniques.…”