China 
Africa 
Explore chapters and articles related to this topic
Digital Image Compression
Published in Edward R. Dougherty, Digital Image Processing Methods, 2020
A k-bit image can be represented by k bit planes, each of the same dimension as the original image. Progressive transmission can be easily achieved by transmitting the bit planes in a sequence, starting with the most significant bit plane and ending with the least significant bit plane. The image reconstructed from the most significant bit plane is a binary image, and additional gray levels are added as more bit planes are received. A lossless reconstruction is possible if all bit planes are used. Since each bit plane is a binary image, binary compression techniques (such as arithmetic coding) can be used to reduce the bit rate [109]. The amount of compression is largest for the most significant bit plane and decreases as one moves through the bit plane sequence. This property allows for large effective compression if the transmission is terminated after receiving the first few bit planes.
Perspectives on Digital Image Watermarking
Published in Roy Subhrajit Sinha, Basu Abhishek, Chattopadhya Avik, Intelligent Copyright Protection for Images, 2019
Roy Subhrajit Sinha, Basu Abhishek, Chattopadhya Avik
The bits from all the pixels together form several bit-planes. For an 8-bit pixel representation, there should be 8 bit-planes in that particular image. The bit-plane concept is shown in Figure 2.3 For any image, the values of higher bit-planes define the perceptual significance of the image, whereas the lower bit-planes describe the smoothing parts. In Figure 2.4, we can see how a grayscale image is portrayed individually by the bit-planes. Here, it is also found that the higher bit-planes are organized in a unique manner, providing a proper shape to the image. But, in the lower bit-planes, bits are appear at random. This is the reason for preferring least significant bit (LSB) planes in imperceptible data hiding practice. This issue is comprehensively discussed in Section 2.2.1.
Advanced Watermarking Techniques
Published in Borra Surekha, Thanki Rohit, Dey Nilanjan, Digital Image Watermarking, 2018
Borra Surekha, Thanki Rohit, Dey Nilanjan
The LSB substitution technique is a basic watermarking technique where less important information or bits of host data are modified by the watermark. In the LSB technique, most significant bits of the watermark are substituted in the LSBs of the host data (Lee and Chen, 2000; Chan and Cheng, 2004; Ramalingam, 2011). In any 8-bit image, the most significant bit plane (i.e., bit plane 7) contains the most important visual information, while the last or least significant bit plane (i.e., bit plane 0) contains no visual information. All other bit planes contribute to various levels of information related to the image. Therefore, the least significant bit plane of the image is chosen for watermarking purposes. For example, to embed an 8-bit gray watermark into a color image, the gray watermark bits are divided into groups of 3 bits, 3 bits, and 2 bits. The first two group’s bits can be inserted into the last three LSB bits of the R channel and G channel. The group of 2 bits can be inserted into the two LSB bits of the B channel. Then, these three channels are combined to generate a watermarked color image. The main advantage of this technique is that after the watermark embedding process, the visual quality of the host data is not much affected by the watermark. Hence, under normal conditions, an average individual cannot see or observe the modifications in the host image where the watermark is inserted. The payload capacity of this technique is almost 100%. This technique is mainly used for copyright authentication of multimedia data. The limitation of this technique is that watermarks embedded in this way are fragile in nature.
A Novel Chaotic Image Encryption Scheme Based on Hash Function and Cyclic Shift
Published in IETE Technical Review, 2019
Xingyuan Wang, Siwei Wang, Na Wei, Yingqian Zhang
According to the minimum unit of encryption, image encryption schemes based on chaos can be divided into two stages: pixel level and bit level. Compared with pixels scrambling, the scrambling in bit level not only changes the information of location but also changes the value of the pixels so that cryptosystems based on the bit level hold higher security [7,8]. Xiang et al. [9] proposed a choice of image encryption scheme which encrypted the high four bits of each pixel but the low four bits remain unchanged. Xu et al. [10] proposed a bit level image encryption algorithm that one bit plane can be permuted into any other bit planes and achieve an excellent encryption performance with only one round. The design theory of our algorithmic is based on transformation in bit level and pixel level simultaneously.
An efficient scheme for the detection of defective parts in fabric images using image processing
Published in The Journal of The Textile Institute, 2023
Toqeer Mahmood, Rehan Ashraf, C. M. Nadeem Faisal
Arivazhagan et al. developed an automated fabric defect detection technique employing multi-scale and multi-orientation-based Gabor wavelet filters for locating the defects in fabric with different yarn thicknesses collected from industry (Arivazhagan et al., 2006). Priya et al. suggested an image processing-based technique to locate fabric defects (Priya et al., 2011). The technique is based on the bit-plane decomposition concept, where the most significant bit planes are utilized to present the solution. However, the proposed solution is ineffective for different types of defects. Jing et al. suggested a fabric defect detection scheme comprising two stages (Jing et al., 2013). Firstly, the researchers extracted texture-based features through Gabor filters to identify the defects in the fabric. To classify the defects, local binary patterns and Tamura algorithm are utilized. However, this method does not perform well for various defects. Çelik et al. presented a discrete wavelet-based technique (Çelik et al., 2014). The Wiener filter is employed to remove noise from the fabric images, followed by soft thresholding and Laplacian operator to accentuate the defective area. Finally, the researchers employed neural networks to classify the defects. The overall accuracy for defect detection and classification reported is 93.4% and 96.3%, respectively. However, this technique is developed to handle only five types of fabric defects. Ng et al. proposed how new kinds of defective and repetitive patterns in fabric images can be examined through image decomposition (Ng et al., 2014). The solution is based on the convex optimization technique that is beneficial for textile and other industries as well.
High-capacity separable reversible data-Hiding method in encrypted images based on block-level encryption and Huffman compression coding
Published in Connection Science, 2021
Kai-Meng Chen, Chin-Chen Chang
Beginning with the 1st bit-plane of the encrypted image, the encrypted bits in the least significant bit (LSB) planes are embedded into the spare room of each high bit-plane to vacate room in the LSB planes of the encrypted image to accommodate the secret data. Fig. 4 shows an example in which the 8th bit-plane and the 7th bit-plane are compressed to vacate room in the LSB plane.