Image captioning is an automated technique that generates textual descriptions of images. It takes an image as input and produces a sentence in English that captures the essential elements and details depicted in the image. This paper explores various image caption-generating techniques based on Deep Neural Network (DNN). These models utilize Convolutional neural networks (CNNs) and their variants, such as InceptionV3, ResNet50, VGG16, and VGG19, to extract visual features from images. The extracted features are then fed into recurrent neural networks (RNNs) such as LSTM and Bi-GRU to generate captions. The paper specifically investigates the impact of different RNN variants on sentence generation, focusing on their ability to capture long-range dependencies. Our evaluation methodology uses BLEU Scores, for assessing the performance of machine translation and image captioning models.