Aiming at the bleed through,page stains,text-like noise on the large area edges and so on,which occurred in the low quality document images,the D -LinkNet framework is improved,and a low quality document image binarizatin al g orithm combining with multiple scale features is proposed.We referred to as MD -LinkNet.The algorithm has two improvements based on the original network.One is to add the remaining multi-core pooling (RMP) module in the middle of the co d ec to extract rich document feature information through four pooling operations; the second is to pass the pooled low-resolution image through DUpsample instea d of bilinear interpolation performs upsampling,not only combines the pixel imag e neighborhood information of the document image,but also to fuses the global a nd local features of the document image to improve the segmentation precision.T he experimental results show that in the 2017and 2018International Document Im age Binarization Competition (DIBCO) dataset,the F-measure of the algorithm re a ches 90.54and 91.42respectively,the results of extensive experiments on many datasets show the robustness of the proposed technique on various types on degra dations in the document images where our technique demonstrates superior perform ance against many other state of art classcal methods.