This research proposes the first approach to integrate the Kolmogorov-Arnold Network (KAN) with various pre-trained Convolutional Neural Network (CNN) models for remote sensing (RS) scene classification using the EuroSAT dataset. The proposed method, named KCN, replaces traditional Multi-Layer Perceptrons (MLPs) with KAN to enhance classification performance. Multiple CNN-based models, including VGG16, MobileNetV2, EfficientNet, ConvNeXt, ResNet101, and Vision Transformer (ViT), were evaluated when paired with KAN. The results showed that KAN achieved high accuracy with fewer training epochs and parameters. Specifically, ConvNeXt paired with KAN achieved 94% accuracy in the first epoch, increasing to 96% and remaining consistent in subsequent epochs. The results indicated that KAN and MLP both achieved similar accuracy, with KAN performing slightly better in later epochs. The EuroSAT dataset was used to provide a robust testbed for investigating whether KAN is suitable for remote sensing classification tasks. Given that KAN is a novel algorithm, there is substantial capacity for further development and optimization, suggesting that KCN offers a promising alternative for efficient image analysis in the RS field. The study highlights the potential of KAN in RS classification tasks, demonstrating its ability to achieve high accuracy with fewer parameters and training epochs. The integration of KAN with ConvNeXt showed that KAN can replace traditional MLPs, achieving satisfactory accuracy. This proves that KCN can successfully combine the strengths of KAN and CNNs, leading to desirable performance for RS scene categorization. Despite these results, the study acknowledges several shortcomings, including the lack of evidence about the interpretability of KAN layers and the need for further research to optimize its performance and integration into diverse remote sensing applications. Future research should address these issues by improving KAN's interpretability and investigating more effective integration solutions for various remote sensing applications. The study concludes that the KAN architecture can replace traditional MLPs for RS scene classification tasks, demonstrating the effectiveness of the KCN approach and its potential to influence the RS field by utilizing the advanced capabilities of KAN and CNN models.This research proposes the first approach to integrate the Kolmogorov-Arnold Network (KAN) with various pre-trained Convolutional Neural Network (CNN) models for remote sensing (RS) scene classification using the EuroSAT dataset. The proposed method, named KCN, replaces traditional Multi-Layer Perceptrons (MLPs) with KAN to enhance classification performance. Multiple CNN-based models, including VGG16, MobileNetV2, EfficientNet, ConvNeXt, ResNet101, and Vision Transformer (ViT), were evaluated when paired with KAN. The results showed that KAN achieved high accuracy with fewer training epochs and parameters. Specifically, ConvNeXt paired with KAN achieved 94% accuracy in the first epoch, increasing to 96% and remaining consistent in subsequent epochs. The results indicated that KAN and MLP both achieved similar accuracy, with KAN performing slightly better in later epochs. The EuroSAT dataset was used to provide a robust testbed for investigating whether KAN is suitable for remote sensing classification tasks. Given that KAN is a novel algorithm, there is substantial capacity for further development and optimization, suggesting that KCN offers a promising alternative for efficient image analysis in the RS field. The study highlights the potential of KAN in RS classification tasks, demonstrating its ability to achieve high accuracy with fewer parameters and training epochs. The integration of KAN with ConvNeXt showed that KAN can replace traditional MLPs, achieving satisfactory accuracy. This proves that KCN can successfully combine the strengths of KAN and CNNs, leading to desirable performance for RS scene categorization. Despite these results, the study acknowledges several shortcomings, including the lack of evidence about the interpretability of KAN layers and the need for further research to optimize its performance and integration into diverse remote sensing applications. Future research should address these issues by improving KAN's interpretability and investigating more effective integration solutions for various remote sensing applications. The study concludes that the KAN architecture can replace traditional MLPs for RS scene classification tasks, demonstrating the effectiveness of the KCN approach and its potential to influence the RS field by utilizing the advanced capabilities of KAN and CNN models.