Skip to content

Commit

Permalink
refactor: 変数名の変更
Browse files Browse the repository at this point in the history
  • Loading branch information
kawano committed Nov 8, 2023
1 parent c55da51 commit 9ed9346
Show file tree
Hide file tree
Showing 2 changed files with 62 additions and 55 deletions.
4 changes: 2 additions & 2 deletions rear_camera_py/Makefile
View file
Original file line number Diff line number Diff line change
Expand Up @@ -101,5 +101,5 @@ make rm:
rm work_image_data/*.png

test-get:
python -m unittest tests/test_calculator.py
# python -m unittest tests/test_get_area_info.py
python -m unittest tests/test_get_area_info.py
# python -m unittest tests/test_calculator.py
113 changes: 60 additions & 53 deletions rear_camera_py/src/get_area_info.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -244,7 +244,7 @@ def pre_process_img(self, course_img) -> None:
return course_img, train_area_img

def check_coordi_about(self,
color_2_img,
color_3_img,
search_color=Color.RED.value,
split_num_height=4,
split_num_width=30,
Expand All @@ -254,7 +254,7 @@ def check_coordi_about(self,
"""色(ブロックやサークル)のおおまかな場所を特定する関数.
Args:
color_2_img (np.ndarray): コース画像
color_3_img (np.ndarray): コース画像
search_color (list): 探すブロックの色
split_num_height (int): 画像の縦を何等分するかの値
split_num_width (int): 画像の横を何等分するかの値
Expand All @@ -265,18 +265,18 @@ def check_coordi_about(self,
mask_coordi: ブロックの位置の大まかな座標(x, y)
develop_img: ブロックの位置を描画した画像(開発用)
"""
mask_height = int(color_2_img.shape[0] / split_num_height)
mask_width = int(color_2_img.shape[1] / split_num_width)
mask_height = int(color_3_img.shape[0] / split_num_height)
mask_width = int(color_3_img.shape[1] / split_num_width)

# (ピクセル数, x, y)
mask_coordi = np.array([0, 0])
max_pixcel = 0

# おおまかなブロックの場所を特定
for x in range(0, int(color_2_img.shape[1]), mask_width):
for x in range(0, int(color_3_img.shape[1]), mask_width):
masks = []
for y in range(split_num_height):
masks.append(color_2_img[mask_height *
masks.append(color_3_img[mask_height *
y:mask_height*(y+1), x:x+mask_width])

# 画素数のカウント
Expand Down Expand Up @@ -493,7 +493,7 @@ def detect_circle(self,
# サークル座標の作成
# 大まかな座標を取得
about_coordi = self.check_coordi_about(
color_2_img=color_5_circle_img,
color_3_img=color_5_circle_img,
search_color=circle_color,
split_num_height=split_num_height,
split_num_width=split_num_width,
Expand Down Expand Up @@ -810,10 +810,14 @@ def predict_block_locate(self, block_coordi) -> (int, int):
_, diff_y_row_1 = Calculator.calc_dis_line_to_coordi(
block_coordi, self.row_verctor_1)

if diff_y_row_0 < diff_y_row_1:
if (diff_y_row_0 is not None and diff_y_row_1 is None) or \
diff_y_row_0 <= diff_y_row_1:
row = 0
else:
elif (diff_y_row_0 is None and diff_y_row_1 is not None) or \
diff_y_row_0 > diff_y_row_1:
row = 1
else:
return None, None

x_arr = np.array([])
# 列
Expand Down Expand Up @@ -889,17 +893,17 @@ def get_area_info(self, isR=True) -> None:

if processed_course_img is None:
if self.develop:
print("[Error]緑のエリアの線分を取得失敗")
print("[Error] 緑のエリアの線分を取得失敗")
return self.course_info_block

# BGR色空間からHSV色空間への変換
game_area_img = cv2.cvtColor(processed_course_img, cv2.COLOR_BGR2HSV)

# 2色画像の作成(赤、青)
color_2_img = np.full_like(game_area_img, Color.WHITE.value, dtype=np.uint8)
self.change_color(game_area_img, color_2_img, self.RED1, Color.RED.value)
self.change_color(game_area_img, color_2_img, self.RED2, Color.RED.value)
self.change_color(game_area_img, color_2_img, self.BLUE, Color.BLUE.value)
# 3色画像の作成(赤、青、白)
color_3_img = np.full_like(game_area_img, Color.WHITE.value, dtype=np.uint8)
self.change_color(game_area_img, color_3_img, self.RED1, Color.RED.value)
self.change_color(game_area_img, color_3_img, self.RED2, Color.RED.value)
self.change_color(game_area_img, color_3_img, self.BLUE, Color.BLUE.value)

# 5色画像の作成（黒以外）
color_5_img = np.full_like(game_area_img, Color.WHITE.value, dtype=np.uint8)
Expand All @@ -925,8 +929,8 @@ def get_area_info(self, isR=True) -> None:
cv2.imwrite(save_path, processed_course_img)
img_count += 1
save_path = os.path.join(self.save_dir_path,
str(img_count)+"_color_2_img.png")
cv2.imwrite(save_path, color_2_img)
str(img_count)+"_color_3_img.png")
cv2.imwrite(save_path, color_3_img)
img_count += 1
save_path = os.path.join(self.save_dir_path,
str(img_count)+"_color_5_img.png")
Expand All @@ -936,7 +940,7 @@ def get_area_info(self, isR=True) -> None:
ブロックの位置を特定する
"""
# ブロック特定用の画像を生成
color_2_under_delete_img = color_2_img.copy()
color_2_under_delete_img = color_3_img.copy()
# 下側削除
for x in range(color_2_under_delete_img.shape[1]):
y = int((self.basis_vector[1] / self.basis_vector[0]) * x * 0.5 + 200)
Expand All @@ -956,9 +960,9 @@ def get_area_info(self, isR=True) -> None:
color_2_under_delete_img, Color.RED.value, 4, 30, 10, block_location_about_img)

# 細かな座標を取得 ([x_max, x_min, y_max, y_min])
block_location_img = color_2_img.copy()
block_location_img = color_3_img.copy()
self.block_coordi_red = self.check_coordi_detail(
color_2_img, about_coordi, Color.RED.value, thre=10, develop_img=block_location_img)
color_3_img, about_coordi, Color.RED.value, thre=10, develop_img=block_location_img)
if self.block_coordi_red is not None:
color_5_circle_img[
self.block_coordi_red[3]:self.block_coordi_red[2],
Expand All @@ -977,7 +981,7 @@ def get_area_info(self, isR=True) -> None:
block_location_about_img)
# 細かな座標を取得
self.block_coordi_blue1 = self.check_coordi_detail(
color_2_img, about_coordi, Color.BLUE.value, thre=20, develop_img=block_location_img)
color_3_img, about_coordi, Color.BLUE.value, thre=20, develop_img=block_location_img)
if self.block_coordi_blue1 is not None:
color_5_circle_img[
self.block_coordi_blue1[3]:self.block_coordi_blue1[2],
Expand All @@ -997,7 +1001,7 @@ def get_area_info(self, isR=True) -> None:
4, 30, 50, block_location_about_img)
# 細かな座標を取得
self.block_coordi_blue2 = self.check_coordi_detail(
color_2_img, about_coordi, Color.BLUE.value, thre=20, develop_img=block_location_img)
color_3_img, about_coordi, Color.BLUE.value, thre=20, develop_img=block_location_img)
if self.block_coordi_blue2 is not None:
color_5_circle_img[
self.block_coordi_blue2[3]:self.block_coordi_blue2[2],
Expand Down Expand Up @@ -1204,9 +1208,8 @@ def get_area_info(self, isR=True) -> None:
row_area_img = color_5_img.copy()

"""
サークルの直線を求める
行の直線を求める
"""
"""行の直線を求める"""
# 3行目(手前)のサークルの直線の式を求める
x = (self.course_info_coordinate[3, :, 0] +
self.course_info_coordinate[3, :, 1])/2
Expand Down Expand Up @@ -1274,7 +1277,9 @@ def get_area_info(self, isR=True) -> None:
color=Color.BLUE.value,
thickness=2)

"""列の直線を求める"""
"""
列の直線を求める
"""
# 0列目(左)のサークルの直線の式を算出する
if np.all(self.course_info_coordinate[3, 0] != 0) and \
np.all(self.course_info_coordinate[2, 0] != 0):
Expand Down Expand Up @@ -1356,35 +1361,39 @@ def get_area_info(self, isR=True) -> None:
cv2.imwrite(save_path, row_area_img)

"""
見つかっていないブロックの座標を確認していく
見つかっていないブロックの座標を確認していく(0,1行目)
"""
"""赤ブロックが見つかっていない場合"""
if self.block_coordi_red is not None and self.block_count_red < self.block_max_count_red:
red_x = np.average(self.block_coordi_red[:2])
red_y = self.block_coordi_red[2]
row, column = self.predict_block_locate(np.array([red_x, red_y]))
if self.course_info_block[row, column] == 0:
self.course_info_block[row, column] = 1
self.block_count_red += 1
if self.develop:
cv2.line(row_area_img,
pt1=(int(red_x), 0), # 始点
pt2=(int(red_x), row_area_img.shape[0]), # 終点
color=Color.BLUE.value,
thickness=2)
cv2.line(row_area_img,
pt1=(0, int(red_y)), # 始点
pt2=(row_area_img.shape[1], int(red_y)), # 終点
color=Color.BLUE.value,
thickness=2)

# 赤ブロックが最後のブロックだった場合、終了
if self.block_count_red + self.block_count_blue >= \
self.block_max_count_red + self.block_max_count_blue or \
(np.all(self.block_coordi_blue1 == 0) and np.all(self.block_coordi_blue2 == 0)):
if row is None:
pass
else:
if self.course_info_block[row, column] == 0:
self.course_info_block[row, column] = 1
self.block_count_red += 1
if self.develop:
print("ブロックすべて発見3")
return self.course_info_block
cv2.line(row_area_img,
pt1=(int(red_x), 0), # 始点
pt2=(int(red_x), row_area_img.shape[0]), # 終点
color=Color.BLUE.value,
thickness=2)
cv2.line(row_area_img,
pt1=(0, int(red_y)), # 始点
pt2=(row_area_img.shape[1], int(red_y)), # 終点
color=Color.BLUE.value,
thickness=2)

# 赤ブロックが最後のブロックだった場合、終了
if self.block_count_red + self.block_count_blue >= \
self.block_max_count_red + self.block_max_count_blue or \
(np.all(self.block_coordi_blue1 == 0) and
np.all(self.block_coordi_blue2 == 0)):
if self.develop:
print("ブロックすべて発見3")
return self.course_info_block

"""青ブロックが見つかっていない場合"""
# """
Expand Down Expand Up @@ -1432,9 +1441,11 @@ def get_area_info(self, isR=True) -> None:
break

row, column = self.predict_block_locate(coordi)
if row is None:
continue

# 手前のサークルが画像上になくて、サークルのベクトルを求めることができなかった場合を考慮
if column is None:
elif column is None:
index = np.where(np.all(self.course_info_coordinate[3] == [0, 0, 0, 0], axis=1))
# 要素が見つかった場合
if len(index[0]) > 0:
Expand Down Expand Up @@ -1487,6 +1498,7 @@ def get_area_info(self, isR=True) -> None:
info.get_area_info(args.isR)
finally:
print(f"最終結果--------------------------------------------------------------")
print(f"image_name : {image_name}")
print(f"block_coordi_red : {info.block_coordi_red}")
print(f"block_coordi_blue1 : {info.block_coordi_blue1}")
print(f"block_coordi_blue2 : {info.block_coordi_blue2}")
Expand All @@ -1497,11 +1509,6 @@ def get_area_info(self, isR=True) -> None:
print(f"course_info_block :\n{info.course_info_block}")
print(f"---------------------------------------------------------------------")

# info.course_info_coordinate
print(f"info.course_info_coordinate[2:, :, 0]: {info.course_info_coordinate[2, :, 0]}")
sorted_indices = np.sort(info.course_info_coordinate[2, :, 0])
print(sorted_indices)

# 処理時間計測用
execute_time = time.time() - start

Expand Down

0 comments on commit 9ed9346

Please sign in to comment.