Histopathology images are normally used to diagnose malignant diseases by examining tissues using a whole slide imaging (WSI) scanner. Though decision support systems analyse histopathological images well in breast cancer detection, there are many issues associated with their choice. The use of different scanners, the usage of various equipment, and the variable stain colouring in the histopathological images are the various contributing factors in the histopathology image-based analysis. The development of efficient colour normalisation techniques for histopathology images is both crucial and difficult. To avoid colour variation artefacts, an adaptive colour deconvolution is proposed in this work. To improve the capacity of colour normalisation, an optimisation is designed to separate the stain. Pathology and computerised decision support systems can benefit from stain separation and colour normalisation of the histology images. On different histopathology image datasets, the quality performance of several colour normalisation techniques is assessed and contrasted in terms of the Pearson Correlation Coefficient (PCC) and the structure similarity index matrix (SSIM). According to our experimental findings, ACD provides better qualitative and quantitative results.