问题
I'm attempting to implement a program originally created by Roger Alsing. I've done quite a bit of research on what other people have implemented. I decided to write my program in python, and use basic triangles as the shapes. When I run the program, it does not show improvement after more generations (The triangles tend to just disappear). I'm assuming something is wrong with my mutate function. Can anyone tell me why its producing less than satisfactory results?
My code:
import random
import copy
from PIL import Image, ImageDraw
optimal = Image.open("mona_lisa.png")
optimal = optimal.convert("RGBA")
size = width, height = optimal.size
num_shapes = 128
generations = 50000
def random_genome():
elements = []
for i in range(num_shapes):
x = (random.randint(0, width), random.randint(0, height))
y = (random.randint(0, width), random.randint(0, height))
z = (random.randint(0, width), random.randint(0, height))
r = random.randint(0, 255)
g = random.randint(0, 255)
b = random.randint(0, 255)
alpha = random.randint(10, 255)
elements.append([x, y, z, r, g, b, alpha])
return elements
def render_daughter(dna):
image = Image.new("RGBA", (width, height), "white")
draw = ImageDraw.Draw(image)
for item in dna:
x = item[0]
y = item[1]
z = item[2]
r = item[3]
g = item[4]
b = item[5]
alpha = item[6]
color = (r, g, b, alpha)
draw.polygon([x, y, z], fill = color)
return image
def mutate(dna):
dna_copy = copy.deepcopy(dna)
shape_index = random.randint(0, len(dna) - 1)
roulette = random.random() * 2
if roulette < 1:
if roulette < 0.25:
dna_copy[shape_index][3] = int(random.triangular(255, dna_copy[shape_index][3]))
elif roulette < 0.5:
dna_copy[shape_index][4] = int(random.triangular(255, dna_copy[shape_index][4]))
elif roulette < 0.75:
dna_copy[shape_index][5] = int(random.triangular(255, dna_copy[shape_index][5]))
elif roulette < 1.0:
dna_copy[shape_index][6] = int(0.00390625 * random.triangular(255, dna_copy[shape_index][6] * 255))
else:
if roulette < 1.25:
dna_copy[shape_index][0] = (int(random.triangular(width, dna_copy[shape_index][0][0])), int(random.triangular(height, dna_copy[shape_index][0][1])))
elif roulette < 1.5:
dna_copy[shape_index][2] = (int(random.triangular(width, dna_copy[shape_index][3][0])), int(random.triangular(height, dna_copy[shape_index][4][1])))
elif roulette < 1.75:
dna_copy[shape_index][3] = (int(random.triangular(width, dna_copy[shape_index][4][0])), int(random.triangular(height, dna_copy[shape_index][5][1])))
return dna_copy
def fitness(original, new):
fitness = 0
for x in range(0, width):
for y in range(0, height):
r1, g1, b1, a1 = original.getpixel((x, y))
r2, g2, b2, a2 = new.getpixel((x, y))
deltaRed = r1 - r2
deltaGreen = g1 - g2
deltaBlue = b1 - b2
deltaAlpha = a1 - a2
pixelFitness = deltaRed + deltaGreen + deltaBlue + deltaAlpha
fitness += pixelFitness
return fitness
def generate():
mother = random_genome()
best_genome = mother
best_fitness = fitness(optimal, render_daughter(best_genome))
for i in range(generations):
daughter = copy.deepcopy(best_genome)
daughter = mutate(daughter)
daughter_fitness = fitness(optimal, render_daughter(daughter))
if daughter_fitness < best_fitness:
best_genome = daughter
best_fitness = daughter_fitness
if i % 50 == 0:
print i
if i % 1000 == 0:
render_daughter(best_genome).save("iterations/output_" + str(i) + ".png")
if __name__ == "__main__":
generate()
The beginning image I am using:
The output image after 1,000 generations:
Output image after 5,000 generations:
回答1:
You're checking wheter the new fitness is smaller than the current fitness:
if daughter_fitness < best_fitness:
The fitness you calculate, however, can be negative:
deltaRed = r1 - r2
deltaGreen = g1 - g2
deltaBlue = b1 - b2
deltaAlpha = a1 - a2
pixelFitness = deltaRed + deltaGreen + deltaBlue + deltaAlpha
fitness += pixelFitness
The various delta* variables can be negative or positive; your test will favour negative deltas, increasing the whiteness of the "best" image (the higher values of r2, g2 etc, the lower the fitness, and the whiter the image, until they are all at 255, 255, 255. I don't know if increasing alpha increases or decreases the transparency).
Thus, you should take the absolute value of the differences:
deltaRed = abs(r1 - r2)
deltaGreen = abs(g1 - g2)
deltaBlue = abs(b1 - b2)
deltaAlpha = abs(a1 - a2)
You could also consider the sum of the square, or square root of the sum of squares (which, basically, turns it into a least-squares fitting routine):
deltaRed = r1 - r2
deltaGreen = g1 - g2
deltaBlue = b1 - b2
deltaAlpha = a1 - a2
pixelFitness = math.sqrt(deltaRed**2 + deltaGreen**2 + deltaBlue**2 + deltaAlpha**2)
fitness += pixelFitness
Finally, I noticed your program doesn't work for me. It's in the second half of your mutate() function, where you assign new values to x, y or z, but use indices above 2. random_genome() shows that you try to access colour values instead, which are integers, and even attempt to index those.
This leads to exceptions, so I don't even know how you could get this program running. It either never ran in the first place, or you didn't properly copy-paste. I've changed that to
if roulette < 1.25:
dna_copy[shape_index][0] = (int(random.triangular(
width, dna_copy[shape_index][0][0])), int(
random.triangular(height, dna_copy[shape_index][0][1])))
elif roulette < 1.5:
dna_copy[shape_index][1] = (int(random.triangular(
width, dna_copy[shape_index][1][0])), int(
random.triangular(height, dna_copy[shape_index][1][1])))
elif roulette < 1.75:
dna_copy[shape_index][2] = (int(random.triangular(
width, dna_copy[shape_index][2][0])), int(
random.triangular(height, dna_copy[shape_index][2][1])))
which seems to do what you want.
来源：https://stackoverflow.com/questions/25134050/incorrect-results-with-genetic-algorithm-image-evolution