An Energy-Efficient Compressive Image Coding for Green Internet of Things (IoT)

Li, Ran; Duan, Xiaomeng; Xu, Li; He, Wei; Li, Yanling

doi:10.3390/s18041231

Cited by 8 publications

(11 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After that, the base JND threshold for each N x N block is computed using the following formula: where i, j are the horizontal and vertical indices of the DCT coefficient at location (i; j) for i; j = 0,…….,N -1 and a = 1.33, b = 0.11, c = 0.18, d = 0.6, e = 0.25. Also, i φ and j φ are defined as: (2) and the spatial frequency of the (i, j)-th DCT coefficient is defined as…”

Section: Methodsmentioning

confidence: 99%

Study of Energy Efficient Images with Just Noticeable Difference Threshold Based on Feature Transform

Soni¹,

Dixit²

2019

IJTSRD

View full text Add to dashboard Cite

A novel method is presented for producing energy-efficient images, i.e., images that consume less electrical energy on energy-adaptive displays, yet have the same, or very similar perceptual quality to their original images. The proposed method relies on the fact the energy consumption of pixels in modern energyadaptive displays like OLED displays are directly proportional to the luminance of the pixels. Hence, in this work to reduce the energy consumption of an image while at the same time preserving its perceptual quality, it is proposed to reduce the luminance of the pixels in the image by one just-noticeable-difference (JND) threshold. To determine the JND thresholds, a Feature Transform based just Noticeable Difference Threshold (FTJNDT) model is developed. In the proposed model, the JND thresholds of each block in the given image are elevated by two non-linear saliency modulation functions using the visual saliency of the block. The parameters of the saliency modulation functions and obtain feature value are estimated through an adaptive optimization framework, which utilizes a state-of-the-art saliency-based objective image quality assessment (IQA) method. To evaluate the proposed methods, a set of subjective experiments were conducted, and the real energy consumption of the produced energy-efficient images were measured by accurate power monitor equipment on an OLED display. The obtained experimental results demonstrated that, on average, the proposed method is able to reduce the energy consumption by about 4.31% while preserving the perceptual quality of the displayed images.

show abstract

Section: Methodsmentioning

confidence: 99%

Study of Energy Efficient Images with Just Noticeable Difference Threshold Based on Feature Transform

Soni¹,

Dixit²

2019

IJTSRD

View full text Add to dashboard Cite

show abstract

“…The same measurement matrix measures the raster scan vector of each block. Since different blocks contain different amounts of valuable information, adaptive block compressed sensing (ABCS) methods [7][8][9][10][11][12][13][14][15][16] have been proposed to make full use of limited measurement resources.…”

Section: Introductionmentioning

confidence: 99%

“…Zhang et al [19] used the standard deviations of blocks as the allocation factor. Many other image features are also effective for ABCS, such as the spatial entropy of blocks [20], the error between blocks [21], the block-based gradient field [11], the block boundary variation [12], and the statistical texture distinctiveness [9]. Moreover, some researchers combined multiple features to allocate the sampling rate of each block.…”

Section: Introductionmentioning

confidence: 99%

Low-Complexity Adaptive Sampling of Block Compressed Sensing Based on Distortion Minimization

Chen

Gong

2022

Sensors

View full text Add to dashboard Cite

Block compressed sensing (BCS) is suitable for image sampling and compression in resource-constrained applications. Adaptive sampling methods can effectively improve the rate-distortion performance of BCS. However, adaptive sampling methods bring high computational complexity to the encoder, which loses the superiority of BCS. In this paper, we focus on improving the adaptive sampling performance at the cost of low computational complexity. Firstly, we analyze the additional computational complexity of the existing adaptive sampling methods for BCS. Secondly, the adaptive sampling problem of BCS is modeled as a distortion minimization problem. We present three distortion models to reveal the relationship between block sampling rate and block distortion and use a simple neural network to predict the model parameters from several measurements. Finally, a fast estimation method is proposed to allocate block sampling rates based on distortion minimization. The results demonstrate that the proposed estimation method of block sampling rates is effective. Two of the three proposed distortion models can make the proposed estimation method have better performance than the existing adaptive sampling methods of BCS. Compared with the calculation of BCS at the sampling rate of 0.1, the additional calculation of the proposed adaptive sampling method is less than 1.9%.

show abstract

“…Compared with the above-mentioned DR techniques, random projection [ 23 , 24 ] is a better choice, since it avoids the model training, but it is still a challenge to store random projections due to the huge dimensionality of text feature. Compressive sensing (CS) [ 25 – 27 ], which has recently been rapidly developing, can be regarded as a random projection technique specially for sparse vectors, and it proves that the perfect recovery of sparse vector can be realized by several random projections. CS retains the advantages of random projection in DR and further overcomes the problem of memory with the help of structural random matrices (SRMs) [ 28 , 29 ], which makes CS a potential DR technique for text classification.…”

Section: Introductionmentioning

confidence: 99%

A Compressive Sensing Model for Speeding Up Text Classification

Shen

Hao

2020

Computational Intelligence and Neuroscience

Self Cite

View full text Add to dashboard Cite

Text classification plays an important role in various applications of big data by automatically classifying massive text documents. However, high dimensionality and sparsity of text features have presented a challenge to efficient classification. In this paper, we propose a compressive sensing- (CS-) based model to speed up text classification. Using CS to reduce the size of feature space, our model has a low time and space complexity while training a text classifier, and the restricted isometry property (RIP) of CS ensures that pairwise distances between text features can be well preserved in the process of dimensionality reduction. In particular, by structural random matrices (SRMs), CS is free from computation and memory limitations in the construction of random projections. Experimental results demonstrate that CS effectively accelerates the text classification while hardly causing any accuracy loss.

show abstract

An Energy-Efficient Compressive Image Coding for Green Internet of Things (IoT)

Cited by 8 publications

References 25 publications

Study of Energy Efficient Images with Just Noticeable Difference Threshold Based on Feature Transform

Study of Energy Efficient Images with Just Noticeable Difference Threshold Based on Feature Transform

Low-Complexity Adaptive Sampling of Block Compressed Sensing Based on Distortion Minimization

A Compressive Sensing Model for Speeding Up Text Classification

Contact Info

Product

Resources

About