The thesis titled "Neural Network for Car Label Recognition" explores the application of deep neural networks for recognizing car brands, including the manufacturer and vehicle model. The study developed and implemented a convolutional neural network (CNN) model for car classification based on images, utilizing deep learning techniques. The main objectives of the thesis are to investigate the potential of neural networks in car label recognition and to enhance the performance of the recognition system through data augmentation, model architecture optimization, and evaluation of various deep learning methods, including transfer learning. The model was tested on a dataset of images captured under different lighting conditions and car orientations. The achieved results were compared with different methods such as ResNet152V2, VGG16, and Xception. The thesis analyzes the impact of different neural network architectures on recognition performance and proposes model improvements to reduce overfitting and increase accuracy on test data.