Poultry meat is a key component of Moroccan cuisine, providing an affordable and nutritious source of protein. With growing demand, ensuring the quality and safety of poultry products has become a critical concern for both producers and consumers. Traditional methods of poultry meat classification, relying on manual inspection, are often subjective, time-consuming, and prone to error. To address these limitations, we use Convolutional Neural Networks (CNNs) for their ability to automatically extract key features like texture and color, enabling accurate classification of poultry types, even in challenging conditions such as minced meat or lighting variations. This study conducts a comparative analysis of four advanced CNN architectures—Mobile Net, ResNet, EfficientNet, and DenseNet—for poultry meat classification. The models are evaluated using key performance metrics, including accuracy, precision, recall, F1-score, and computational efficiency, on a dataset of high-resolution poultry meat images categorized by quality, freshness, and type. The findings provide insights into the strengths and weaknesses of each model, balancing accuracy and computational cost. This research contributes to the field of food quality assessment using deep learning and offers practical recommendations for implementing CNN-based solutions in the Moroccan poultry industry, enhancing the efficiency and reliability of meat classification for producers, retailers, and consumers.