Fashion trend analysis using computer vision is gradually becoming a global trend. It helps the apparel industry to identify the choice of their potential customers and focus on how to increase sales based on the study. In this research we propose a Convolutional Neural Network (CNN) based model for developing an intelligent apparel image classification system. We have prepared our own dataset containing a total of 1000 images collected from social networking sites i.e. Google, Facebook, Instagram, and LinkedIn. The system can classify items into ten categories such as shirt, t-shirt, Punjabi, sweater, blazer, saree, salwar kameez, western tops, gown, and party wear. Our proposed model employs 7 CNN architectures, including MobileNetV2, MobileNetV3, InceptionV3, EfficientNetB0, EfficientNetB3, VGG19, and DenseNet201 where EfficientNetB3 achieves the highest accuracy (86%), followed by VGG19 (85%). This study will contribute to personalized fashion experiences and assist apparel industries in targeting specific customers' profiles.