In this article, we propose Deep Transfer Learning (DTL) Model for recognizing covid-19 from chest x-ray images. The latter is less expensive, easily accessible to populations in rural and remote areas. In addition, the device for acquiring these images is easy to disinfect, clean and maintain. The main challenge is the lack of labeled training data needed to train convolutional neural networks. To overcome this issue, we propose to leverage Deep Transfer Learning architecture pre-trained on ImageNet dataset and trained Fine-Tuning on a dataset prepared by collecting normal, COVID-19, and other chest pneumonia X-ray images from different available databases. We take the weights of the layers of each network already pre-trained to our model and we only train the last layers of the network on our collected COVID-19 image dataset. In this way, we will ensure a fast and precise convergence of our model despite the small number of COVID-19 images collected. In addition, for improving the accuracy of our global model will only predict at the output the prediction having obtained a maximum score among the predictions of the seven pre-trained CNNs. The proposed model will address a three-class classification problem: COVID-19 class, pneumonia class, and normal class. To show the location of the important regions of the image which strongly participated in the prediction of the considered class, we will use the Gradient Weighted Class Activation Mapping (Grad-CAM) approach. A comparative study was carried out to show the robustness of the prediction of our model compared to the visual prediction of radiologists. The proposed model is more efficient with a test accuracy of 98%, an f1 score of 98.33%, an accuracy of 98.66% and a sensitivity of 98.33% at the time when the prediction by renowned radiologists could not exceed an accuracy of 63.34% with a sensitivity of 70% and an f1 score of 66.67%.

Three-Dimensional Hahn moments are performant tool in the domain of image processing applications and pattern classification. In this work, we propose a new method for computing the Three-Dimensional Hahn moments. This method is based on matrix multiplication and symmetry property to decrease the complexity and computational time for volumetric image reconstruction. Experimental results showed that the proposed method is very efficient in terms of computation time, but also in terms of volumetric image reconstruction capability.

In this paper, we propose a new set of separable two-dimensional discrete orthogonal moments called Krawtcouk-Tchebichef's moments. This set of moments is based on the bivariate discrete orthogonal polynomials defined from the product of Krawtchouk and Tchebichef discrete orthogonal polynomials with one variable. We also present a novel set of Krawtchouk-Tchebichef invariant moments. These invariant moments are derived algebraically from the geometric invariant moments and their computation is accelerated using an image representation scheme. The performance of these invariant moments used as pattern features for a pattern classification is compared with Tchebichef and Krawtchouk invariant moments.

This paper present a full on-chip and area efficient low dropout voltage regulator(LDO), exploiting the nested miller compensation with active capacitor (NMCAC) to eliminate the external capacitor and active feedback resistors to reduce the chip area. The external capacitor is removed allowing for greater power system integration for system on-chip applications. The idea has applied to stabilize a 1.6 V, 50 mA Low dropout regulator. Using the proposed techniques the regulator LDO works with a supply voltage as low as 1.8 V and provides a load current 50 mA with a dropout voltage of 200 mV and output variation 4 mV when a full load step 0-50 mA is applied. It designed in 0.18 μm CMOS technology.

This paper presents A low noise, high PSRR low-dropout regulator for low-cost portable electronics. The proposed LDO uses an Operational Transconductance Amplifier (OTA) with Miller R-C compensation to ensure a stable phase margin greater than 62°. The active chip of the proposed regulator is only 150×680 um2. The design was simulated and lyouted in Cadence using mixed-signal 90 nm 2P9M CMOS process, the simulation results show that this LDO output voltage can achieves line regulation of less than 0.10% and load regulation of better than 0.25% and a low quiescent current of only 90uA and ultra-low noise of only 65 nV/SqrtHz. Moreover, it can achieve a PSRR of -52 dB and -64 dB at 1 and 10 kHz, respectively. The input voltage is ranged from 2.70 to 5 V for a load current 100 mA and an output voltage of 1.5 V.

The standard least squared distance method of fitting a line to a set of data points is known to be unreliable when the random noise in the input is significant compared with the data correlated to the line itself. Here, we present a new statistical clustering method based on Legendre moment theory and maximum entropy principle for line fitting in a noisy image. We propose a new approach for estimating the underlying probability density function (p.d.f.) of the data set. The p.d.f. is expanded in terms of Legendre polynomials by means of the Legendre moments. The order of the expansion is selected according to the maximum entropy principle. Then, the points corresponding to the maxima of the p.d.f. will be the true points of the line to be extracted by a chaining algorithm. This approach is directly generalized to multidimensional data. The proposed algorithm was successfully applied to real and simulated noisy line images, with comparison to some well-known methods.