Intersatellite optical wireless communication (IsOWC) systems are crucial for global communication networks. However, their performance is significantly affected by external disturbance, which degrade efficiency and impact the eye diagram image. To address these challenges, a novel technique using a denoising convolutional autoencoder (DCAE) model is proposed to enhance noisy eye diagram images. The augmented noisy temporal and amplitude eye diagram images generated from a quadrature amplitude modulation-based IsOWC system. These images incorporate noise with standard deviations ranging from 0.01 to 0.1. In the DCAE framework, an encoder compresses the noisy eye diagram image into a lower-dimensional representation, while a decoder reconstructs a clean version, minimizing the mean squared error (MSE) or reconstruction error between the noisy input and the denoised output. Remarkably, the proposed model achieves an MSE of 0.02. The DCAE model effectively reduces noise up to a certain threshold (SD ~0.04), its performance deteriorates at higher noise levels (SD = 0.05). On the other hand, the inter_lanczos4 interpolation method, widely used for enhancing image quality by resizing, demonstrates the lowest average MSE at approximately 16 compared to other methods. Such advancements are crucial for applications in terrestrial, non-terrestrial, deep space, and underwater communication systems.