You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

301 lines
9.5 KiB
Python

import numpy as np
import cv2
import time
from functools import wraps
def find_the_adjacent_boxes(boxes: list):
list_index = list()
for i, a in enumerate(boxes):
for j, b in enumerate(boxes):
if i == j:
continue
if j in list_index:
continue
ax, ay, aw, ah = single_xyxy2cxcywh(a)
bx, by, bw, bh = single_xyxy2cxcywh(b)
dis = l2((ax, ay), (bx, by))
if dis < 2 * aw or dis < 2 * bw:
list_index.append(i)
list_index.append(j)
list_index = set(list_index)
return list(list_index)
def l2(p1, p2):
x0, y0 = p1
x1, y1 = p2
return np.sqrt((x0 - x1) ** 2 + (y0 - y1) ** 2)
def single_xyxy2cxcywh(box):
x1, y1, x2, y2 = box
w = x2 - x1
h = y2 - y1
x = x1 + w / 2
y = y1 + h / 2
return x, y, w, h
def sigmoid(x):
return 1 / (1 + np.exp(-x))
def xywh2xyxy(x):
# Convert [x, y, w, h] to [x1, y1, x2, y2]
y = np.copy(x)
y[:, 0] = x[:, 0] - x[:, 2] / 2 # top left x
y[:, 1] = x[:, 1] - x[:, 3] / 2 # top left y
y[:, 2] = x[:, 0] + x[:, 2] / 2 # bottom right x
y[:, 3] = x[:, 1] + x[:, 3] / 2 # bottom right y
return y
def process(input, mask, anchors, size):
anchors = [anchors[i] for i in mask]
grid_h, grid_w = map(int, input.shape[0:2])
box_confidence = sigmoid(input[..., 4])
box_confidence = np.expand_dims(box_confidence, axis=-1)
box_class_probs = sigmoid(input[..., 5:])
box_xy = sigmoid(input[..., :2]) * 2 - 0.5
col = np.tile(np.arange(0, grid_w), grid_h).reshape(-1, grid_w)
row = np.tile(np.arange(0, grid_h).reshape(-1, 1), grid_w)
col = col.reshape(grid_h, grid_w, 1, 1).repeat(3, axis=-2)
row = row.reshape(grid_h, grid_w, 1, 1).repeat(3, axis=-2)
grid = np.concatenate((col, row), axis=-1)
box_xy += grid
box_xy *= (int(size[1] / grid_h), int(size[0] / grid_w))
box_wh = pow(sigmoid(input[..., 2:4]) * 2, 2)
box_wh = box_wh * anchors
box = np.concatenate((box_xy, box_wh), axis=-1)
return box, box_confidence, box_class_probs
def filter_boxes(boxes, box_confidences, box_class_probs, box_threshold, nms_threshold):
"""Filter boxes with box threshold. It's a bit different with origin yolov5 post process!
# Arguments
boxes: ndarray, boxes of objects.
box_confidences: ndarray, confidences of objects.
box_class_probs: ndarray, class_probs of objects.
# Returns
boxes: ndarray, filtered boxes.
classes: ndarray, classes for boxes.
scores: ndarray, scores for boxes.
"""
boxes = boxes.reshape(-1, 4)
box_confidences = box_confidences.reshape(-1)
box_class_probs = box_class_probs.reshape(-1, box_class_probs.shape[-1])
_box_pos = np.where(box_confidences >= box_threshold)
boxes = boxes[_box_pos]
box_confidences = box_confidences[_box_pos]
box_class_probs = box_class_probs[_box_pos]
class_max_score = np.max(box_class_probs, axis=-1)
classes = np.argmax(box_class_probs, axis=-1)
_class_pos = np.where(class_max_score * box_confidences >= box_threshold)
boxes = boxes[_class_pos]
classes = classes[_class_pos]
scores = (class_max_score * box_confidences)[_class_pos]
return boxes, classes, scores
def nms_boxes(boxes, scores, nms_threshold):
"""Suppress non-maximal boxes.
# Arguments
boxes: ndarray, boxes of objects.
scores: ndarray, scores of objects.
# Returns
keep: ndarray, index of effective boxes.
"""
x = boxes[:, 0]
y = boxes[:, 1]
w = boxes[:, 2] - boxes[:, 0]
h = boxes[:, 3] - boxes[:, 1]
areas = w * h
order = scores.argsort()[::-1]
keep = []
while order.size > 0:
i = order[0]
keep.append(i)
xx1 = np.maximum(x[i], x[order[1:]])
yy1 = np.maximum(y[i], y[order[1:]])
xx2 = np.minimum(x[i] + w[i], x[order[1:]] + w[order[1:]])
yy2 = np.minimum(y[i] + h[i], y[order[1:]] + h[order[1:]])
w1 = np.maximum(0.0, xx2 - xx1 + 0.00001)
h1 = np.maximum(0.0, yy2 - yy1 + 0.00001)
inter = w1 * h1
ovr = inter / (areas[i] + areas[order[1:]] - inter)
inds = np.where(ovr <= nms_threshold)[0]
order = order[inds + 1]
keep = np.array(keep)
return keep
def restore_bound_box(boxes: list, ratio: tuple, pad_size: tuple):
if len(boxes) > 0:
pad_width, pad_height = pad_size
boxes_array = np.asarray(boxes)
# print(boxes_array)
boxes_array[:, 0] = (boxes_array[:, 0] - pad_width) / ratio[0]
boxes_array[:, 1] = (boxes_array[:, 1] - pad_height) / ratio[1]
boxes_array[:, 2] = (boxes_array[:, 2] - pad_width) / ratio[0]
boxes_array[:, 3] = (boxes_array[:, 3] - pad_height) / ratio[1]
# print(boxes_array)
boxes = boxes_array.tolist()
return boxes
def letterbox(im, new_shape=(640, 640), color=(0, 0, 0)):
# Resize and pad image while meeting stride-multiple constraints
shape = im.shape[:2] # current shape [height, width]
if isinstance(new_shape, int):
new_shape = (new_shape, new_shape)
# Scale ratio (new / old)
r = min(new_shape[0] / shape[0], new_shape[1] / shape[1])
# Compute padding
ratio = r, r # width, height ratios
new_unpad = int(round(shape[1] * r)), int(round(shape[0] * r))
dw, dh = new_shape[1] - new_unpad[0], new_shape[0] - new_unpad[1] # wh padding
dw /= 2 # divide padding into 2 sides
dh /= 2
if shape[::-1] != new_unpad: # resize
im = cv2.resize(im, new_unpad, interpolation=cv2.INTER_LINEAR)
top, bottom = int(round(dh - 0.1)), int(round(dh + 0.1))
left, right = int(round(dw - 0.1)), int(round(dw + 0.1))
im = cv2.copyMakeBorder(im, top, bottom, left, right, cv2.BORDER_CONSTANT, value=color) # add border
return im, ratio, (dw, dh)
def cost(tag=''):
try:
'''
:param tag: 装饰器的参数
'''
from loguru import logger
def wrapper(fn):
@wraps(fn)
def wrapper_use_time(*args, **kw):
'''
函数传参方式 fn(arg1, arg2, param1=a, param2=b)
:param args: 位置参数(arg1, arg2, ....) 传参方式 fn(arg1, arg2, ...)
:param kw: 字典参数{param1=a, param2=b, .....} 传参方式 fn(param1=a, param2=b, ....)
:return:
'''
t1 = time.time()
try:
res = fn(*args, **kw)
except Exception as e:
logger.error(f"@use_time %s(%s) execute error" % (fn.__name__, tag))
return None
else:
t2 = time.time()
logger.info(f"{tag}@UseTime: {t2 - t1}")
return res
return wrapper_use_time
return wrapper
except Exception as err:
print(err)
def align_box(imgs, bbox, size=96, scale_factor=1.0, center_bias=0, borderValue=(0, 0, 0)):
bias_x = (-1 + 2 * np.random.sample()) * center_bias
bias_y = (-1 + 2 * np.random.sample()) * center_bias
b_x1, b_y1, b_x2, b_y2 = bbox
cx, cy = (b_x1 + b_x2) // 2, (b_y1 + b_y2) // 2
w = b_x2 - b_x1
h = b_y2 - b_y1
cx += w * bias_x
cy += h * bias_y
base_r = max(w, h)
j_x = 0
j_y = 0
j_r = 0
base_r += j_r
r = int(base_r / 2 * scale_factor)
cy -= int(base_r * 0)
cx += j_x
cy += j_y
x1, y1, x2, y2 = cx - r, cy - r, cx + r, cy + r
x3, y3 = cx - r, cy + r
_x1, _y1, _x2, _y2, _x3, _y3 = [0, 0, size, size, 0, size]
src = np.array([x1, y1, x2, y2, x3, y3], dtype=np.float32).reshape(3, 2)
sv = np.asarray([[b_x1, b_y1, 1], [b_x2, b_y2, 1]])
dst = np.array([_x1, _y1, _x2, _y2, _x3, _y3], dtype=np.float32).reshape(3, 2)
assert src.dtype == np.float32
assert dst.dtype == np.float32
assert src.shape == (3, 2)
assert dst.shape == (3, 2)
mat = cv2.getAffineTransform(src, dst)
p = sv.dot(mat.T).reshape(-1)
if type(imgs) == list:
imgs = [cv2.warpAffine(img, mat, (size, size), borderValue=borderValue) for img in imgs]
else:
imgs = cv2.warpAffine(imgs, mat, (size, size), borderValue=borderValue)
return imgs, p, mat
def get_rotate_crop_image(img, points):
'''
img_height, img_width = img.shape[0:2]
left = int(np.min(points[:, 0]))
right = int(np.max(points[:, 0]))
top = int(np.min(points[:, 1]))
bottom = int(np.max(points[:, 1]))
img_crop = img[top:bottom, left:right, :].copy()
points[:, 0] = points[:, 0] - left
points[:, 1] = points[:, 1] - top
'''
assert len(points) == 4, "shape of points must be 4*2"
img_crop_width = int(
max(
np.linalg.norm(points[0] - points[1]),
np.linalg.norm(points[2] - points[3])))
img_crop_height = int(
max(
np.linalg.norm(points[0] - points[3]),
np.linalg.norm(points[1] - points[2])))
pts_std = np.float32([[0, 0], [img_crop_width, 0],
[img_crop_width, img_crop_height],
[0, img_crop_height]])
points = points.astype(np.float32)
# print(points.shape, pts_std.shape)
M = cv2.getPerspectiveTransform(points, pts_std)
dst_img = cv2.warpPerspective(
img,
M, (img_crop_width, img_crop_height),
borderMode=cv2.BORDER_REPLICATE,
flags=cv2.INTER_CUBIC)
dst_img_height, dst_img_width = dst_img.shape[0:2]
if dst_img_height * 1.0 / dst_img_width >= 1.5:
dst_img = np.rot90(dst_img)
return dst_img