def fitness(self):
"""Evaluates fitness value.
Fitness value is calculated as sum of dissimilarity measures between each adjacent pieces.
"""
if self._fitness is None:
fitness_value = 1 / self.FITNESS_FACTOR
# For each two adjacent pieces in rows
for i in range(self.rows):
for j in range(self.columns - 1):
ids = (self[i][j].id, self[i][j + 1].id)
fitness_value += ImageAnalysis.get_dissimilarity(ids, orientation="LR")
# For each two adjacent pieces in columns
for i in range(self.rows - 1):
for j in range(self.columns):
ids = (self[i][j].id, self[i + 1][j].id)
fitness_value += ImageAnalysis.get_dissimilarity(ids, orientation="TD")
self._fitness = self.FITNESS_FACTOR / fitness_value
return self._fitness
print(self._fitness)
def fitness(self):
"""Evaluates fitness value.
Fitness value is calculated as sum of dissimilarity measures between each adjacent pieces.
这个方法会在第一次调用时被执行，懒初始化fitness，结果是适应度因子除以空间距离，（结果越大适应度越好）
"""
if self._fitness is None:
fitness_value = 1 / self.FITNESS_FACTOR
# For each two adjacent pieces in rows
# 累加所有左右关系的碎片的空间距离
for i in range(self.rows):
for j in range(self.columns - 1):
ids = (self[i][j].id, self[i][j + 1].id)
fitness_value += ImageAnalysis.get_dissimilarity(
ids, orientation="LR")
# For each two adjacent pieces in columns
# 累加所有上下关系的碎片的空间距离
for i in range(self.rows - 1):
for j in range(self.columns):
ids = (self[i][j].id, self[i + 1][j].id)
fitness_value += ImageAnalysis.get_dissimilarity(
ids, orientation="TD")
self._fitness = self.FITNESS_FACTOR / fitness_value
return self._fitness
def start_evolution(self, verbose):
print("=== Pieces: {}\n".format(len(self._pieces)))
if verbose:
plot = Plot(self._image)
ImageAnalysis.analyze_image(self._pieces)
fittest = None
best_fitness_score = float("-inf")
termination_counter = 0
for generation in range(self._generations):
print_progress(generation,
self._generations - 1,
prefix="=== Solving puzzle: ")
new_population = []
# Elitism
elite = self._get_elite_individuals(elites=self._elite_size)
new_population.extend(elite)
selected_parents = roulette_selection(self._population,
elites=self._elite_size)
for first_parent, second_parent in selected_parents:
crossover = Crossover(first_parent, second_parent)
crossover.run()
child = crossover.child()
new_population.append(child)
fittest = self._best_individual()
if fittest.fitness <= best_fitness_score:
termination_counter += 1
else:
best_fitness_score = fittest.fitness
if termination_counter == self.TERMINATION_THRESHOLD:
print("\n\n=== GA terminated")
print("=== There was no improvement for {} generations".format(
self.TERMINATION_THRESHOLD))
return fittest
self._population = new_population
if verbose:
plot.show_fittest(
fittest.to_image(),
"Generation: {} / {}".format(generation + 1,
self._generations))
return fittest
def _get_buddy_piece(self, piece_id, orientation):
first_buddy = ImageAnalysis.best_match(piece_id, orientation)
second_buddy = ImageAnalysis.best_match(
first_buddy, complementary_orientation(orientation))
if second_buddy == piece_id:
for edge in [
parent.edge(piece_id, orientation)
for parent in self._parents
]:
if edge == first_buddy:
return edge
def _get_buddy_piece(self, piece_id, orientation):
first_buddy = ImageAnalysis.best_match(piece_id, orientation)
second_buddy = ImageAnalysis.best_match(
first_buddy, complementary_orientation(orientation))
# 如果两块互相为匹配度最高的块，则表示很有可能是相邻块
if second_buddy == piece_id:
for edge in [
parent.edge(piece_id, orientation)
for parent in self._parents
]:
if edge == first_buddy:
# 如果父母中有任意一方有最佳匹配块，则返回
return edge
def best_adjoin(self, piece_size):
pieces = np.reshape(self.pieces, (self.rows, self.columns))
empty_image = np.zeros(
(piece_size, piece_size, pieces[0][0].shape()[2]))
empty_piece = Piece(empty_image, -1)
for row in range(self.rows):
for col in range(self.columns):
if row == 0:
if col == 0 and ImageAnalysis.in_range(pieces[row][col].id, "D", pieces[row + 1][col].id) \
and ImageAnalysis.in_range(pieces[row][col].id, "R", pieces[row][col + 1].id):
continue
if col < self.columns - 1 and ImageAnalysis.in_range(pieces[row][col].id, "D",
pieces[row + 1][col].id) \
and ImageAnalysis.in_range(pieces[row][col].id, "R", pieces[row][col + 1].id):
continue
if col == self.columns - 1 and ImageAnalysis.in_range(pieces[row][col].id, "D",
pieces[row + 1][col].id) \
and ImageAnalysis.in_range(pieces[row][col].id, "L", pieces[row][col - 1].id):
continue
if 0 < row < self.rows - 1:
if col == 0 and ImageAnalysis.in_range(pieces[row][col].id, "D", pieces[row + 1][col].id) \
and ImageAnalysis.in_range(pieces[row][col].id, "R", pieces[row][col + 1].id):
continue
if col < self.columns - 1 and ImageAnalysis.in_range(pieces[row][col].id, "D",
pieces[row + 1][col].id) \
and ImageAnalysis.in_range(pieces[row][col].id, "R", pieces[row][col + 1].id):
continue
if col == self.columns - 1 and ImageAnalysis.in_range(pieces[row][col].id, "D",
pieces[row + 1][col].id) \
and ImageAnalysis.in_range(pieces[row][col].id, "L", pieces[row][col - 1].id):
continue
if row == self.rows - 1:
if col == 0 and ImageAnalysis.in_range(pieces[row][col].id, "T", pieces[row - 1][col].id) \
and ImageAnalysis.in_range(pieces[row][col].id, "R", pieces[row][col + 1].id):
continue
if col < self.columns - 1 and ImageAnalysis.in_range(pieces[row][col].id, "R",
pieces[row][col + 1].id) \
and ImageAnalysis.in_range(pieces[row][col].id, "T", pieces[row - 1][col].id):
continue
if col == self.columns - 1 and ImageAnalysis.in_range(pieces[row][col].id, "L",
pieces[row][col - 1].id) \
and ImageAnalysis.in_range(pieces[row][col].id, "T", pieces[row - 1][col].id):
continue
# if row == 0:
# if col == 0 and ImageAnalysis.best_match(pieces[row][col].id, "D") == pieces[row + 1][col].id \
# and ImageAnalysis.best_match(pieces[row][col].id, "R") == pieces[row][col + 1].id:
# continue
# if col < self.columns - 1 and ImageAnalysis.best_match(pieces[row][col].id, "D") == pieces[row + 1][col].id \
# and ImageAnalysis.best_match(pieces[row][col].id, "R") == pieces[row][col + 1].id \
# and ImageAnalysis.best_match(pieces[row][col].id, "L") == pieces[row][col - 1].id:
# continue
# if col == self.columns - 1 and ImageAnalysis.best_match(pieces[row][col].id, "D") == pieces[row + 1][col].id \
# and ImageAnalysis.best_match(pieces[row][col].id, "L") == pieces[row][col - 1].id:
# continue
# if 0 < row < self.rows - 1:
# if col == 0 and ImageAnalysis.best_match(pieces[row][col].id, "D") == pieces[row + 1][col].id \
# and ImageAnalysis.best_match(pieces[row][col].id, "T") == pieces[row - 1][col].id \
# and ImageAnalysis.best_match(pieces[row][col].id, "R") == pieces[row][col + 1].id:
# continue
# if col < self.columns - 1 and ImageAnalysis.best_match(pieces[row][col].id, "T") == pieces[row - 1][col].id \
# and ImageAnalysis.best_match(pieces[row][col].id, "D") == pieces[row + 1][col].id \
# and ImageAnalysis.best_match(pieces[row][col].id, "L") == pieces[row][col - 1].id \
# and ImageAnalysis.best_match(pieces[row][col].id, "R") == pieces[row][col + 1].id:
# continue
# if col == self.columns - 1 and ImageAnalysis.best_match(pieces[row][col].id, "T") == pieces[row - 1][col].id \
# and ImageAnalysis.best_match(pieces[row][col].id, "D") == pieces[row + 1][col].id \
# and ImageAnalysis.best_match(pieces[row][col].id, "L") == pieces[row][col - 1].id:
# continue
# if row == self.rows - 1:
# if col == 0 and ImageAnalysis.best_match(pieces[row][col].id, "T") == pieces[row - 1][col].id \
# and ImageAnalysis.best_match(pieces[row][col].id, "R") == pieces[row][col + 1].id:
# continue
# if col < self.columns - 1 and ImageAnalysis.best_match(pieces[row][col].id, "R") == pieces[row][col + 1].id \
# and ImageAnalysis.best_match(pieces[row][col].id, "L") == pieces[row][col - 1].id \
# and ImageAnalysis.best_match(pieces[row][col].id, "T") == pieces[row - 1][col].id:
# continue
# if col == self.columns - 1 and ImageAnalysis.best_match(pieces[row][col].id, "L") == pieces[row][col - 1].id \
# and ImageAnalysis.best_match(pieces[row][col].id, "T") == pieces[row - 1][col].id:
# continue
pieces[row][col] = empty_piece
self.penalize_image = pieces
return image_helpers.assemble_image(
[each.image for each in pieces.flatten()], self.rows, self.columns)
def start_evolution(self, verbose):
print("=== Pieces: {}\n".format(len(self._pieces)))
if verbose:
plot = Plot(self._image)
ImageAnalysis.analyze_image(self._pieces)
fittest = None
best_fitness_score = float("-inf")
termination_counter = 0
for generation in range(self._generations):
print_progress(generation,
self._generations - 1,
prefix="=== Solving puzzle: ")
new_population = []
# Elitism
# 取适应度最高的两个图片
elite = self._get_elite_individuals(elites=self._elite_size)
new_population.extend(elite)
# 从种群中随机选择popultation - elite_size个父母
selected_parents = roulette_selection(self._population,
elites=self._elite_size)
# 通过父母生成子代，加入到new_population中
for first_parent, second_parent in selected_parents:
# 交叉互换，生成子代
crossover = Crossover(first_parent, second_parent)
crossover.run()
child = crossover.child()
# child.mutate()
new_population.append(child)
# 从上一代中选出适应度最高的一个
fittest = self._best_individual()
fittest.mutate()
# min_fitness = 0
# for index in range(len(new_population)):
# if new_population[index].fitness < new_population[min_fitness].fitness:
# min_fitness = index
# fittest.clear_fitness()
# if fittest.fitness > new_population[min_fitness].fitness:
# new_population[min_fitness] = fittest
print("old_fittest : ", fittest.fitness, end="")
# image = fittest.to_image()
# rightImage = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)
# cv2.imwrite("temp_image_" + str(generation) + ".jpg", rightImage)
best_adjoin = fittest.best_adjoin(self._piece_size)
rightImage = cv2.cvtColor(best_adjoin, cv2.COLOR_RGB2BGR)
cv2.imwrite("temp_image_best_adjoin_" + str(generation) + ".jpg",
rightImage)
# penalisze = fittest.penalize()
# print(" new_fittest : ", fittest.fitness)
# rightImage = cv2.cvtColor(penalize, cv2.COLOR_RGB2BGR)
# cv2.imwrite("temp_image_penalize_" + str(generation) + ".jpg", rightImage)
# 如果上一代最佳比历史最佳好，则termination_counter += 1,否则替换
if fittest.fitness < best_fitness_score:
termination_counter += 1
else:
best_fitness_score = fittest.fitness
termination_counter = 0
if termination_counter % 4 == 2:
predicate = Individual(fittest.pieces,
fittest.rows,
fittest.columns,
shuffle=False)
predicate.penalize_image = fittest.penalize_image
# 处理局部最优
predicate.manually_select()
# predicate.shuffle_assembling()
print("predicate_fitness : %s " % str(predicate.fitness))
image = predicate.to_image()
rightImage = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)
cv2.imwrite("predicate_image_" + str(generation) + ".jpg",
rightImage)
for index in range(len(new_population)):
if new_population[index].fitness < predicate.fitness:
new_population[index] = predicate
break
# 如果连续十代都没有更优子代，则退出
if termination_counter == self.TERMINATION_THRESHOLD:
print("\n\n=== GA terminated")
print("=== There was no improvement for {} generations".format(
self.TERMINATION_THRESHOLD))
return fittest
self._population = new_population
if verbose:
plot.show_fittest(
fittest.to_image(),
"Generation: {} / {}".format(generation + 1,
self._generations))
return fittest