Input: X-ray Chest Images for different types (Normal, COVID, Pneumonia) Output: Features extracted from RESCOVIDTCNNet EWT (Pre-processing Step) Step1: The X-ray chest images are passed to the EWT for de-nosing and pre-processing. Step2: Two Dimensional Littlewood-Paley EWT is presented To deal with 2D images. Step3: The details and the approximation coefficients based on 2D Dimensional Littlewood-Paley EWT are obtained using the following equations: $W_I^{\epsilon LP}\left(n,t\right)=F_2^{\ast }(F_2\left(I\right)\left(\omega \right)\overline{F_2\left({\psi }_n\right)\left(\omega \right)}$ $W_I^{\epsilon LP}\left(0,t\right)=F_2^{\ast }(F_2\left(I\right)\left(\omega \right)\overline{F_2\left({\mathrm{\varnothing }}_n\right)\left(\omega \right)}$ Step4: The X-ray chest images are reconstructed from the detailed coefficients of EWT. Resnet50 (Feature Extraction) Step5: The Pre-processed EWT X-ray chest images are passed to the transfer learning model known by Resnet50. Step6: The images are passed to 5 main blocks. The first block consists of convolutional and max-pooling layers. Step7: The second block consists of 9 convolutional layers, and the third block consists of 12 convolutional layers. Step8: The output of the third block is based on the fourth block, which consists of 18 convolutional layers, and the output is passed to the fifth block, which consists of convolutional layers. Step9: Then, an average pooling and fully connected layers are applied to the output of the fifth block. Step10: A deep residual learning is presented to create shortcut connections by mapping the layers to residual known by $H\left(x\right)$. Step11: The nonlinear layers are mapped to another type of mapping function defined by: $F\left(x\right)$: =$H\left(x\right)-x$ TCN (Feature Extraction) Step12: The output of the fully connected layer in Resnet50 is passed to the temporal convolutional network (TCN). Step13: The sequence input layer of TCN accepts the output of Resnet50, and it is passed to four main residual blocks. Step14: The output of the residual blocks is passed to the fully connected layer. Step15: The features of the fully connected layer are input to a set of classifiers which are ANN and SVM.