def main():
# load obstacles
template_files = os.listdir(r"images/obstacle_images")
for index, fileName in enumerate(template_files):
template_files[index] = r'images/obstacle_images/' + fileName
dinosaur_path = './images/dinosaur.PNG'
# Create image processing object to use detection with
imgProcess = ImageProcess(template_files, dinosaur_path)
print('focus game')
time.sleep(2)
i = 0
gameOver = False
while gameOver != -1:
img = ScreenCapture.get_screen(top=172,
left=-1543,
width=600,
height=125,
delay=0)
gameOver = imgProcess.get_distance(img)
if gameOver:
print(gameOver)
i += 1
def main():
# load obstacles
template_files = os.listdir("images/obstacle_images")
for index, fileName in enumerate(template_files):
template_files[index] = "images/obstacle_images/" + fileName
# Create image processing object to use detection with
imgProcess = ImageProcess(template_files, "images/dinosaur.PNG")
#get image and find distance to all obstacles
img = ScreenCapture.get_screen(
top = 172, left = -1543, width = 600,
height = 125, delay = 0
)
cv2.imshow('test', img)
cv2.waitKey(0)
cv2.destroyAllWindows()
res = imgProcess.get_distance(img, drawRect = False)
print(res)
# raise IndexError
#get score of the run
score_img = ScreenCapture.get_screen(
top=142, left= -1009, width=65, height=20, delay=0
)
# imgProcess.show_image(cv2.cvtColor(score_img, cv2.COLOR_BGRA2GRAY))
score = imgProcess.get_score(score_img, show_score=True)
print(int(score))
def __init__(self, create_info=True):
self.score = None
# Create image processing object to use detection with
template_files = os.listdir("images/obstacle_images")
for index, fileName in enumerate(template_files):
template_files[index] = "images/obstacle_images/" + fileName
dinosaur_image_path = r"images/dinosaur.PNG"
self.game_vision = ImageProcess(template_files, dinosaur_image_path)
self.obs_names = self.game_vision.get_obs_names()
try:
self.obs_names.remove('game_over')
except ValueError:
pass
self.decisionGenes = {name: None for name in self.obs_names}
# print(self.decisionGenes)
for key in self.decisionGenes.keys():
self.decisionGenes[key] = np.empty([650, 100], dtype=object)
logging.basicConfig(filename='play_excepts.log',
filemode='w',
format='%(name)s - %(levelname)s - %(message)s')
if create_info:
self.create_genetic_info()
def button_spit_num(self, bool):
image = self.paint.getPixMap()
log.info_out(image)
self.image_spit = ImageProcess(image, self.spit_Box.currentIndex() + 1)
if len(self.image_spit) == 0:
QMessageBox.warning(self, "waring", "无法正确分割数字，或者无数字")
return
point = self.image_spit.get_rect_point()
for x, y, x1, y1 in point:
self.paint.paintLine(x, y, x1 - x, y1 - y)
def button_recognize(self, bool):
image = self.paint.getPixMap()
self.image_spit = ImageProcess(image, self.spit_Box.currentIndex() + 1)
if len(self.image_spit) == 0:
QMessageBox.warning(self, "waring", "无法正确分割数字，或者无数字")
return
num = []
for index, (image) in enumerate(self.image_spit):
# 自动膨胀和腐蚀
if self.AutoED.isChecked():
kernel = np.ones((2, 2), np.uint8)
image = cv2.dilate(image, kernel)
kernel = np.ones((1, 1), np.uint8)
image = cv2.erode(image, kernel)
if self.BinProcess.isChecked():
image[image[:, :] < 100] = 0
image[image[:, :] > 100] = 255
# 显示测试图片预览
if self.StudyPreView.isChecked():
plt.figure(index)
plt.imshow(image, cmap="gray")
plt.show()
# 数据转化为Torch格式
data = torch.tensor([[image]]).float()
# 开始识别
out = self.net(data)
# 找到最大值
num.append(torch.max(out.data, 1)[1][0].item())
# text控件显示结果
s = ""
for i in num:
s = s + str(i)
self.lineEdit.setText(s)
def myUI(self, w1, h1, w2, h2, name_input, name_output):
'''
given function performs the following tasks:
1. read the image in BGR format
2. convert w1, h1, w2, h2 window size in respective pixel format
3. convert BGR image into Luv image
4. find histogram of entire image on Luv domain,
where L is in range of given window
5. convert Luv image into BGR image
6. write output image
'''
# 1. read the image in BGR format
myIO = MyIO()
bgrImg = myIO.readImage(name_input)
# debug
print("bgrImg =\n {}".format(bgrImg))
# debug -ends
# debug
myIO.showImage(bgrImg, "BGR Image")
# debug -ends
# 2. convert w1, h1, w2, h2 window size in respective pixel format
W1, H1, W2, H2 = myIO.windowsSizeMapping(inputImage = bgrImg,\
w1 = w1, h1=h1,\
w2 = w2, h2=h2)
# debug
print("W1 = {}, H1={}, W2={}, H2={}".format(W1, H1, W2, H2))
# debug -ends
# 3. convert BGR image into Luv image
colorProcess=ColorProcess()
LuvImg = colorProcess.bgrToLuv(bgrImg = bgrImg)
# debug
print("-----------------------------------------------------")
print("\nLuvImg = \n{}".format(LuvImg))
# debug -ends
#4. find histogram of entire image on Luv domain,
# where L is in range of given window
imageProcess = ImageProcess()
HELuvImg = imageProcess.histogramEqualizationInLuv(LuvImg, W1, H1, W2, H2)
# debug
print("-----------------------------------------------------")
print("HELuvImg = \n{}".format(HELuvImg))
# debug -ends
# 5. convert Luv image into BGR image
HEBGRImage = colorProcess.LuvToBGR(LuvImage = HELuvImg)
# debug
myIO.showImage(HEBGRImage, "Histogram Equalized BGR Image")
cv2.waitKey(0)
# debug -ends
# debug
print("-----------------------------------------------------")
print("HEBGRImage =\n {}".format(HEBGRImage))
# debug -ends
#6. write output image
myIO.writeImage(outputImage = HEBGRImage, name_output = name_output)
class mainWindow(QMainWindow, Ui_MainWindow):
def __init__(self):
super(mainWindow, self).__init__()
self.setupUi(self)
icon = QtGui.QIcon()
icon.addPixmap(QtGui.QPixmap("UI/ico/windows.png"), QtGui.QIcon.Normal,
QtGui.QIcon.Off)
self.setWindowIcon(icon)
self.setWindowTitle("手写数字识别")
self.paint = Painter(750, 520)
self.gridLayout.addWidget(self.paint)
self.net = Net()
self.net.load_state_dict(torch.load("net/parameters.pt"))
self.image_spit = None
self.spinBox_px.setValue(10)
with open("UI/css/mainwindow.css", "r") as file_css:
self.setStyleSheet(file_css.read())
def button_spit_num(self, bool):
image = self.paint.getPixMap()
log.info_out(image)
self.image_spit = ImageProcess(image, self.spit_Box.currentIndex() + 1)
if len(self.image_spit) == 0:
QMessageBox.warning(self, "waring", "无法正确分割数字，或者无数字")
return
point = self.image_spit.get_rect_point()
for x, y, x1, y1 in point:
self.paint.paintLine(x, y, x1 - x, y1 - y)
def button_clear(self, bool):
self.paint.clearWidget()
self.lineEdit.clear()
def button_recognize(self, bool):
image = self.paint.getPixMap()
self.image_spit = ImageProcess(image, self.spit_Box.currentIndex() + 1)
if len(self.image_spit) == 0:
QMessageBox.warning(self, "waring", "无法正确分割数字，或者无数字")
return
num = []
for index, (image) in enumerate(self.image_spit):
# 自动膨胀和腐蚀
if self.AutoED.isChecked():
kernel = np.ones((2, 2), np.uint8)
image = cv2.dilate(image, kernel)
kernel = np.ones((1, 1), np.uint8)
image = cv2.erode(image, kernel)
if self.BinProcess.isChecked():
image[image[:, :] < 100] = 0
image[image[:, :] > 100] = 255
# 显示测试图片预览
if self.StudyPreView.isChecked():
plt.figure(index)
plt.imshow(image, cmap="gray")
plt.show()
# 数据转化为Torch格式
data = torch.tensor([[image]]).float()
# 开始识别
out = self.net(data)
# 找到最大值
num.append(torch.max(out.data, 1)[1][0].item())
# text控件显示结果
s = ""
for i in num:
s = s + str(i)
self.lineEdit.setText(s)
def button_withdraw(self, bool):
self.paint.withDraw()
def button_clear_alltemp(self, bool):
self.paint.disImage()
def button_writeTemp(self, bool):
self.paint.mergeTemp2Image()
def button_preView(self, bool):
self.paint.ProcessPic(self.YH_Box.currentIndex())
def spin_num_change(self, num):
self.paint.paintSize = num
self.paint.update()
class Player:
def __init__(self, create_info=True):
self.score = None
# Create image processing object to use detection with
template_files = os.listdir("images/obstacle_images")
for index, fileName in enumerate(template_files):
template_files[index] = "images/obstacle_images/" + fileName
dinosaur_image_path = r"images/dinosaur.PNG"
self.game_vision = ImageProcess(template_files, dinosaur_image_path)
self.obs_names = self.game_vision.get_obs_names()
try:
self.obs_names.remove('game_over')
except ValueError:
pass
self.decisionGenes = {name: None for name in self.obs_names}
# print(self.decisionGenes)
for key in self.decisionGenes.keys():
self.decisionGenes[key] = np.empty([650, 100], dtype=object)
logging.basicConfig(filename='play_excepts.log',
filemode='w',
format='%(name)s - %(levelname)s - %(message)s')
if create_info:
self.create_genetic_info()
def jump(self, time, keypressMut): # press up key for `time` seconds
if keypressMut.acquire(blocking=False):
# print("action is jump, time = ", time)
keyDown("up")
keyup_timer = threading.Timer(time, keyUp, args=("up", ))
lock_release_timer = threading.Timer(time, keypressMut.release)
keyup_timer.start()
lock_release_timer.start()
else:
# print('another action has not released')
pass
def duck(self, time, keypressMut): # press the down key for `time` seconds
if keypressMut.acquire(blocking=False):
# print('action is duck, time = ', time)
keyDown("down")
keyup_timer = threading.Timer(time, keyUp, args=("down", ))
lock_release_timer = threading.Timer(time, keypressMut.release)
keyup_timer.start()
lock_release_timer.start()
else:
# print('another action has not released')
pass
def do_nothing(self, time, keypressMut):
if keypressMut.acquire(blocking=False):
# print('action is nothing, time = ', time)
lock_release_timer = threading.Timer(time, keypressMut.release)
lock_release_timer.start()
else:
# print('another action has not released')
pass
def create_genetic_info(self):
#ground obstacles can only be jumped over
#bird can be ducked or jumped over
ground_obs_actions = [self.jump, self.do_nothing]
bird_obs_actions = [self.jump, self.do_nothing, self.duck]
for key in self.decisionGenes.keys():
if 'bird' not in key:
for i in range(len(self.decisionGenes[key])):
for j in range(len(self.decisionGenes[key][i])):
randSleep = random.uniform(0.1, 1.5)
self.decisionGenes[key][i][j] = [
random.choice(ground_obs_actions), randSleep
]
else:
for i in range(len(self.decisionGenes[key])):
for j in range(len(self.decisionGenes[key][i])):
randSleep = random.uniform(0.1, 1.5)
self.decisionGenes[key][i][j] = [
random.choice(bird_obs_actions), randSleep
]
def noneTest(self):
for key in self.decisionGenes:
for x_list in self.decisionGenes[key]:
for y in x_list:
if None in y:
print('key: ', key)
@staticmethod
def copy_player(individual):
new_player = Player(create_info=False)
for key in individual.decisionGenes:
new_player.decisionGenes[key] = copy.deepcopy(
individual.decisionGenes[key])
return new_player
def play(self):
keypressMut = threading.Lock()
pattern = re.compile('_\d+')
game_over = False
time.sleep(1) #game actually resets
press("space") # start game
time.sleep(0.5) #game starts
print("game started ")
while not game_over:
# game starts. find image and take action
img = ScreenCapture.get_screen(top=172,
left=-1543,
width=600,
height=125,
delay=0.4)
res = self.game_vision.get_distance(img)
if res:
# print(res)
try:
obstacle, distance = res
match = pattern.search(obstacle)
obs_name = obstacle[:match.span()[0]]
action, wait = self.decisionGenes[obs_name][distance[0]][
distance[1]]
if action == None:
print(action, wait, distance[0], distance[1])
raise IndexError
action(wait, keypressMut)
#res = -1
except TypeError as e:
game_over = True
score_img = ScreenCapture.get_screen(top=142,
left=-1009,
width=65,
height=20,
delay=0)
try:
score = int(self.game_vision.get_score(score_img))
#Score could not be converted to int
#Issue with OCR software
except ValueError as e:
with open('./errors/score_read.log', 'a+') as scoreFail:
scoreFail.write(str(e))
t = time.localtime()
timestamp = time.strftime('%b-%d-%Y_%H%M', t)
FILE_NAME = ("score_fail-" + timestamp)
cv2.imwrite(r'./errors' + FILE_NAME + '.bmp', score_img)
score = 42
print('game done', score, '\n')
return score
def __len__(self):
return 95100