At present, the whole world is infected by COVID-19. It targets affecting the respiratory system and worsens for people with other health complications like diabetes, cardiovascular diseases, cancer, lung disorder, and so on. The availability of test kits is not adequate and symptoms of COVID-19 similar to pneumonia are deadly, which claims millions of people. The COVID-19 test kits are time-consuming and even reduce the detection rate. Therefore, in the current study, an automatic CT image classification technique for COVID-19 and Non-COVID patient identification is proposed. In this paper, a StackedAlexNet-19 convolution network for automatic classification of lung CT images is proposed. The proposed StackedAlexNet-19 model consists of different pre-trained methods like ResNet 101, Xception, NASNet, MobileNet, and InceptionV3. Based on the pre-trained model, input CT images are processed and integrated for the detection of abnormalities in COVID-19 CT images of patients. The StackAleNet-19 model is evaluated and comparatively examined with the existing techniques. The dataset for processing consists of 1359 CT images composed of COVID, non-COVID, and other infections. The validation range is set as 50 for each case with a total value of 150 and the network is trained with CT images of 1069 for classification. The analysis of results expressed that StackAlexNet-19 exhibits higher accuracy, sensitivity, and specificity value of 93.67%, 0.93, and 0.97 respectively. The proposed StackAlexNet classification technique achieves an accuracy of 93.67%. The developed model provides improved accuracy than the existing techniques. The StackAlexNet-19 facilitates the intervention of COVID-19 without any human intervention.