<p>As the primary cause of death for women across the globe, breast cancer plays a major role in global mortality statistics. Timely detection and proper treatment are crucial in minimizing the impact of breast cancer. The analysis of breast tissue through histopathology is fundamental in diagnosing breast cancer. Early identification is key to improving survival outcomes and patients' overall well-being. To address this, an effective model known as the Pyramid Kronecker Forward Harmonic Network (PyKFHNet ) is developed for detecting breast cancer from histopathological images. Primarily, the input histopathological image is forwarded into the image denoising stage, and it is performed utilizing a linear Filter. Thereafter, blood cell segmentation is performed by employing RefineNet. Afterwards, feature extraction is performed to extract the features, which consist of shape features like area, solidity, eccentricity, perimeter, major axis length and Discrete Cosine Transform (DCT) with Entropy-based Local Binary Pattern (ELBP). Finally, breast cancer detection is conducted using PyKFHNet that is the incorporation of PyramidNet and Deep Kronecker network (DKN). This novel combination enables hierarchical feature learning and efficient feature representation, improving detection accuracy compared to conventional approaches. Experimental results demonstrate that PyKFHNet achieves a maximum accuracy of 90.719%, sensitivity of 90.533%, and specificity of 91.348%, highlighting its effectiveness in automated breast cancer detection.</p>