本文实例为大家分享了python扫描线填充算法,供大家参考,具体内容如下
介绍
1.用水平扫描线从上到下扫描由点线段构成的多段构成的多边形。
2.每根扫描线与多边形各边产生一系列交点。将这些交点按照x坐标进行分类,将分类后的交点成对取出,作为两个端点,以所填的色彩画水平直线。
3.多边形被扫描完毕后,填色也就完成。
数据结构
活性边表:
新边表:
代码(使用数组)
import numpy as np from PIL import Image from PIL import ImageDraw from PIL import ImageFont array = np.ndarray((660, 660, 3), np.uint8) array[:, :, 0] = 255 array[:, :, 1] = 255 array[:, :, 2] = 255 for i in range(0,660,1): array[i,330]=(0,0,0) for j in range(0,660,1): array[330,j]=(0,0,0) def creat_Net(point, row, y_min,y_max ): Net = [([ ] * y_max ) for i in range(y_max )] point_count = point.shape[0] for j in range(0, point_count): x = np.zeros(10) first = int(min(point[(j+1)%point_count][1] , point[j][1])) x[1] = 1/((point[(j+1)%point_count][1]-point[j][1])/(point[(j+1)%point_count][0]-point[j][0])) # x 的增量 x[2] = max(point[(j+1)%point_count][1] , point[j][1]) if(point[(j+1)%point_count][1] < point[j][1]): x[0] = point[(j+1)%point_count][0] else: x[0] = point[j][0] Net[first].append(x) return Net def draw_line(i,x ,y ): for j in range(int(x),int(y)+1): array[330-i,j+330]=(20,20,20) def polygon_fill(point): y_min = np.min(point[:,1]) y_max = np.max(point[:,1]) Net = creat_Net(point, y_max - y_min + 1, y_min, y_max) x_sort = [] * 3 for i in range(y_min, y_max): x = Net[i] if(len(x) != 0): for k in x : x_sort.append(k) x_image = [] * 3 for cell in x_sort: x_image.append(cell[0]) x_image.sort() if(len(x_image) >= 3 and x_image[0]==x_image[1] and x_image[2]>x_image[1]): x_image[1] = x_image[2] draw_line(i, x_image[0], x_image[1]) linshi = [] * 3 for cell in x_sort: if cell[2] > i: cell[0] += cell[1] linshi.append(cell) x_sort = linshi[:] x_image = [] * 3 for cell in x_sort: x_image.append(cell[0]) x_image.sort() draw_line(i, x_image[0],x_image[1]) def main(): point = [[55,40], [100,80], [100,160],[55,180], [10,160], [10,80]] point = np.array(point) polygon_fill( point ) image = Image.fromarray(array) image.save('saomao.jpg') image.show() if __name__ == "__main__": main()
实例:
以上就是本文的全部内容,希望对大家的学习有所帮助,也希望大家多多支持亿速云。
免责声明:本站发布的内容(图片、视频和文字)以原创、转载和分享为主,文章观点不代表本网站立场,如果涉及侵权请联系站长邮箱:is@yisu.com进行举报,并提供相关证据,一经查实,将立刻删除涉嫌侵权内容。