Relationship between entropy and SNR changes in image enhancement

Krbcová, Zuzana; Kukal, Jaromír

doi:10.1186/s13640-017-0232-z

Cited by 6 publications

(5 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This could lead to different mounting conditions and consequently affect the performance of the optical tracking system. Furthermore, differences in contrast and SNR could potentially affect the results obtained by the objective evaluation 26,27 . It is noteworthy that in certain cases (shown in Figure 8), reduction in artifacts became obvious in the PMC‐ON image only following close inspection.…”

Section: Discussionmentioning

confidence: 99%

Quantitative evaluation of prospective motion correction in healthy subjects at 7T MRI

Sciarra

Mattern

Yakupov

et al. 2021

Magnetic Resonance in Med

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 99%

Quantitative evaluation of prospective motion correction in healthy subjects at 7T MRI

Sciarra

Mattern

Yakupov

et al. 2021

Magnetic Resonance in Med

View full text Add to dashboard Cite

show abstract

“…Entropy measures are widely used in image processing for image enhancements such as de-noising and image restoration/reconstruction using de-convolution [14][15]- [16]. Maximum Entropy (ME) is used to measure the maximum amount of information in images.…”

Section: B Maximum Entropy In Image Datamentioning

confidence: 99%

Examining and Mitigating Kernel Saturation in Convolutional Neural Networks using Negative Images

Gowdra

Sinha

MacDonell

2020

IECON 2020 the 46th Annual Conference of the IEEE Industrial Electronics Society

View full text Add to dashboard Cite

Neural saturation in Deep Neural Networks (DNNs) has been studied extensively, but remains relatively unexplored in Convolutional Neural Networks (CNNs). Understanding and alleviating the effects of convolutional kernel saturation is critical for enhancing CNN models classification accuracies. In this paper, we analyze the effect of convolutional kernel saturation in CNNs and propose a simple data augmentation technique to mitigate saturation and increase classification accuracy, by supplementing negative images to the training dataset. We hypothesize that greater semantic feature information can be extracted using negative images since they have the same structural information as standard images but differ in their data representations. Varied data representations decrease the probability of kernel saturation and thus increase the effectiveness of kernel weight updates. The two datasets selected to evaluate our hypothesis were CIFAR-10 and STL-10 as they have similar image classes but differ in image resolutions thus making for a better understanding of the saturation phenomenon. MNIST dataset was used to highlight the ineffectiveness of the technique for linearly separable data. The ResNet CNN architecture was chosen since the skip connections in the network ensure the most important features contributing the most to classification accuracy are retained. Our results show that CNNs are indeed susceptible to convolutional kernel saturation and that supplementing negative images to the training dataset can offer a statistically significant increase in classification accuracies when compared against models trained on the original datasets. Our results present accuracy increases of 6.98% and 3.16% on the STL-10 and CIFAR-10 datasets respectively.

show abstract

“…Entropy measures are widely used in image processing for image enhancements such as de-noising and image restoration/reconstruction using deconvolutions [6], [8], [9]. Hartley Entropy (HART) or Maximum Entropy (ME) is the de-facto standard for measuring the maximum amount of information in applications of digital image processing.…”

Section: B Maximum Entropy In Image Datamentioning

confidence: 99%

“…ME and SNR measures are commonly used in digital image processing [6], [7] to precisely determine quantitative measurements for signal information present in the input data. The motivation for our novel research is based on the widespread use of SNR and ME values in image processing [8], [9], to enhance and restore degraded images. Through our experiments and analyses we show that the use of these measures to tailor CNN models yields statistically significant and improved classification performance.…”

Section: Introductionmentioning

confidence: 99%

Examining convolutional feature extraction using Maximum Entropy (ME) and Signal-to-Noise Ratio (SNR) for image classification

Gowdra

Sinha

MacDonell

2020

IECON 2020 the 46th Annual Conference of the IEEE Industrial Electronics Society

View full text Add to dashboard Cite

Convolutional Neural Networks (CNNs) specialize in feature extraction rather than function mapping. In doing so they form complex internal hierarchical feature representations, the complexity of which gradually increases with a corresponding increment in neural network depth. In this paper, we examine the feature extraction capabilities of CNNs using Maximum Entropy (ME) and Signal-to-Noise Ratio (SNR) to validate the idea that, CNN models should be tailored for a given task and complexity of the input data. SNR and ME measures are used as they can accurately determine in the input dataset, the relative amount of signal information to the random noise and the maximum amount of information respectively. We use two well known benchmarking datasets, MNIST and CIFAR-10 to examine the information extraction and abstraction capabilities of CNNs. Through our experiments, we examine convolutional feature extraction and abstraction capabilities in CNNs and show that the classification accuracy or performance of CNNs is greatly dependent on the amount, complexity and quality of the signal information present in the input data. Furthermore, we show the effect of information overflow and underflow on CNN classification accuracies. Our hypothesis is that the feature extraction and abstraction capabilities of convolutional layers are limited and therefore, CNN models should be tailored to the input data by using appropriately sized CNNs based on the SNR and ME measures of the input dataset.

show abstract

Relationship between entropy and SNR changes in image enhancement

Cited by 6 publications

References 15 publications

Quantitative evaluation of prospective motion correction in healthy subjects at 7T MRI

Quantitative evaluation of prospective motion correction in healthy subjects at 7T MRI

Examining and Mitigating Kernel Saturation in Convolutional Neural Networks using Negative Images

Examining convolutional feature extraction using Maximum Entropy (ME) and Signal-to-Noise Ratio (SNR) for image classification

Contact Info

Product

Resources

About