MB-LBP特征-matlab

将图像分成一个个小块（Block），每个小块再分为一个个的小区域（类似于HOG中的cell），小区域内的灰度平均值作为当前小区域的灰度值，与周围小区域灰度进行比较形成LBP特征，生成的特征称为MB-LBP，Block大小为3*3，则小区域的大小为1，就是原始的LBP特征

---

matlab源码

function imglbp = getMultiScaleBlockLBPFeature(img, scale)
    imgSize = size(img);
    if numel(imgSize) > 2
        imgG = rgb2gray(img);
    else
        imgG = img;
    end
    [rows, cols] = size(imgG);
    rows=int16(rows);
    cols=int16(cols);

    cellSize = floor(scale / 3);
    offset = floor(cellSize / 2);
    im = uint8(zeros(rows-2*offset, cols-2*offset));

    for i=offset:(rows-offset-1)
        for j=offset:(cols-offset-1)
            temp = 0;
            for m=-offset:offset
                for n=-offset:offset
                    temp = temp + int16(imgG(i+n+1,j+m+1));
                end
            end
            temp = floor(temp / (cellSize * cellSize));

            temp = uint8(temp);
            im(i - offset + 1, j - offset + 1) = temp;
        end
    end
    %imshow(im)
    imglbp = getOriginLBPFeature(im);
end

---

效果图

原图

scale = 9