Cracks in asphalt pose significant safety risks to roads and highways, necessitating effective and efficient inspection methods. Manual inspection approaches are not only costly but also prone to errors. To address these challenges, this paper introduced an integrated model for automated road crack classification. The methodology comprised four key steps: image segmentation, noise reduction, feature extraction, and crack classification. In the initial stages, the paper presented a novel forest optimization algorithm (FOA) tailored for optimizing the Otsu thresholding method. Leveraging a forest-based optimization approach, this algorithm harnessed the collective decision-making power of multiple trees to identify the optimal threshold value for image segmentation. Subsequently, a hybrid feature extraction approach was proposed, combining histograms of oriented gradients (HOG) and Harris corner detection. HOG captures texture information through the analysis of local gradients, while Harris corner detection identifies distinctive features. The fusion of these techniques enhanced the discriminative power of the extracted features, providing a robust image representation for subsequent classification tasks. To fine-tune the hyperparameters of the k-nearest neighbors (kNN) classifier, the paper incorporated Bayesian optimization. This approach efficiently explored the hyperparameter space, identifying optimal parameter settings that enhance the classification performance of the model. By combining the optimized kNN classifier with the extracted features, the integrated model aimed to achieve accurate image classification for segmented regions. Experimental results indicate the efficacy of the proposed hybrid model, demonstrating the highest accuracy at 98.10%. This outcome signified the model's effectiveness in precisely detecting and classifying cracks in asphalt roads. The achieved accuracy, coupled with the systematic integration of novel algorithms and approaches, validated the potential of the proposed model to significantly improve the efficiency of crack detection processes. The integrated model showcased promise for automating road crack classification, reducing reliance on manual inspection, and providing accurate results crucial for road safety and maintenance.