A processed function is used to transform an original image into a processed one in traditional image processing. This paper introduce a approach that processes images without any required for transformation functions, utilizing artificial neural networks based on the Heb-bian learning rule. The main contribution of this method is its ability to learn from the visual features of images rather than algorithmic approach. The proposed framework features an adaptive Artificial Neural Networks topology that adjusts in response to variations in the local characteristics of the input image. Moreover, this paper presents an adaptive Hebbian learning algorithm designed to for image filtering processes by the application of neural network . Traditional image filtering methods often struggle with noise and detail preservation, leading to sub optimal results in various applications. This approach based the principles of Hebbian learning, which focuses on the correlation of neural activations, to dynamically adjust the filtering parameters based on the input data characteristics. The proposed method( Adaptive Hebbian Neural Network) demonstrates increase the performance compared to traditional algorithms with 0.0019 and 0.9309 for the Least Mean Squares and Delta rule algorithms, respectively. It usful for advanced image processing applications, offering a powerful tool for researchers and practitioners in the field.