This paper presents an enhanced version of the VGG19 pre-trained model, referred to as E-VGG19, for the detection and classification of skin cancer. The E-VGG19 model incorporates max pooling and dense layers to improve feature extraction and classification accuracy. The study compares the performance of the E-VGG19 model with other pre-trained models such as ResNet152v2, InceptionResNetV2, DenseNet201, and ResNet50. The training dataset consists of skin lesion images, including both malignant and benign cases. Various machine learning methods, including Linear Support Vector Machine (SVM), k-Nearest Neighbors (KNN), Decision Tree (DT), Logistic Regression (LR), and Support Vector Machine (SVM), are employed to classify the images. The results demonstrate that combining the E-VGG19 model with traditional classifiers significantly improves the overall classification accuracy for skin cancer detection. The performance metrics, including recall, F1 score, precision, sensitivity, and accuracy, are used to evaluate the models. The experiment results provide valuable insights into the effectiveness of various models and classifiers for accurate and efficient skin cancer detection. The research contributes to the ongoing efforts to create automated technologies for early skin cancer detection, which can help healthcare professionals and individuals identify potential skin cancer cases at an early stage, leading to more timely and effective treatments.This paper presents an enhanced version of the VGG19 pre-trained model, referred to as E-VGG19, for the detection and classification of skin cancer. The E-VGG19 model incorporates max pooling and dense layers to improve feature extraction and classification accuracy. The study compares the performance of the E-VGG19 model with other pre-trained models such as ResNet152v2, InceptionResNetV2, DenseNet201, and ResNet50. The training dataset consists of skin lesion images, including both malignant and benign cases. Various machine learning methods, including Linear Support Vector Machine (SVM), k-Nearest Neighbors (KNN), Decision Tree (DT), Logistic Regression (LR), and Support Vector Machine (SVM), are employed to classify the images. The results demonstrate that combining the E-VGG19 model with traditional classifiers significantly improves the overall classification accuracy for skin cancer detection. The performance metrics, including recall, F1 score, precision, sensitivity, and accuracy, are used to evaluate the models. The experiment results provide valuable insights into the effectiveness of various models and classifiers for accurate and efficient skin cancer detection. The research contributes to the ongoing efforts to create automated technologies for early skin cancer detection, which can help healthcare professionals and individuals identify potential skin cancer cases at an early stage, leading to more timely and effective treatments.