This work evaluates Sloth-Enhanced Contrast Limited Adaptive Histogram Equalization (S-CLAHE) effectiveness in enhancing underwater image classification performance. S-CLAHE aims to enhance the contrast and quality of underwater images by integrating the exploration strategy inspired by sloths into the traditional CLAHE algorithm. S-CLAHE is compared with DeepSeaNet and MCANet, the existing deep learning algorithms, in a number of ways, such as True Positive Rate, True Negative Rate, Precision, Matthews Correlation Coefficient, Classification Accuracy, and F-Measure. With better classification accuracy and dependability, the findings reveal that S-CLAHE routinely beats the other methods. These results show the possibilities of S-CLAHE as a useful tool for improving underwater picture classification tasks, so applicable in marine biology, archaeology, and exploration.