from ctypes import *
import math
import random
import cv2
import os
import numpy as np
import shutil
def clip(image,overlap,nh,nw):
h,w,c=image.shape
y_step=nh-overlap
x_step=nw-overlap
rois=[]
for row in range(0,h,y_step):
for col in range(0,w,x_step):
if row==0 and col==0:
rois.append(image[row:row+nh,col:col+nw,:])
if row==0 and col>0:
if (col+nw)>=w:
ww=w-col
else:
ww=nw
if ww<=0:
continue
if ww >= 416:
rois.append(image[row:row+nh,col:col+ww,:])
if row>0 and col==0:
if (row+nw)>=h:
hh=h-row
else:
hh=nh
if hh<=0:
continue
if hh >=416:
rois.append(image[row:row+hh,col:col+nw,:])
if row>0 and col >0:
if (col+nw)>=w:
ww=w-col
else:
ww=nw
if (row+nh)>=h:
hh=h-row
else:
hh=nh
if ww<=0 or hh<=0:
continue
if ww>=416 and hh>=416:
rois.append(image[row:row+hh,col:col+ww,:])
return np.asarray(rois)
def sample(probs):
s = sum(probs)
probs = [a/s for a in probs]
r = random.uniform(0, 1)
for i in range(len(probs)):
r = r - probs[i]
if r <= 0:
return i
return len(probs)-1
def c_array(ctype, values):
arr = (ctype*len(values))()
arr[:] = values
return arr
class BOX(Structure):
fields = [(“x”, c_float),
(“y”, c_float),
(“w”, c_float),
(“h”, c_float)]
class DETECTION(Structure):
fields = [(“bbox”, BOX),
(“classes”, c_int),
(“prob”, POINTER(c_float)),
(“mask”, POINTER(c_float)),
(“objectness”, c_float),
(“sort_class”, c_int)]
class IMAGE(Structure):
fields = [(“w”, c_int),
(“h”, c_int),
(“c”, c_int),
(“data”, POINTER(c_float))]
class METADATA(Structure):
fields = [(“classes”, c_int),
(“names”, POINTER(c_char_p))]
#lib = CDLL("/home/pjreddie/documents/darknet/libdarknet.so", RTLD_GLOBAL)
lib = CDLL(“libdarknet.so”, RTLD_GLOBAL)
lib.network_width.argtypes = [c_void_p]
lib.network_width.restype = c_int
lib.network_height.argtypes = [c_void_p]
lib.network_height.restype = c_int
predict = lib.network_predict
predict.argtypes = [c_void_p, POINTER(c_float)]
predict.restype = POINTER(c_float)
set_gpu = lib.cuda_set_device
set_gpu.argtypes = [c_int]
make_image = lib.make_image
make_image.argtypes = [c_int, c_int, c_int]
make_image.restype = IMAGE
get_network_boxes = lib.get_network_boxes
get_network_boxes.argtypes = [c_void_p, c_int, c_int, c_float, c_float, POINTER(c_int), c_int, POINTER(c_int)]
get_network_boxes.restype = POINTER(DETECTION)
make_network_boxes = lib.make_network_boxes
make_network_boxes.argtypes = [c_void_p]
make_network_boxes.restype = POINTER(DETECTION)
free_detections = lib.free_detections
free_detections.argtypes = [POINTER(DETECTION), c_int]
free_ptrs = lib.free_ptrs
free_ptrs.argtypes = [POINTER(c_void_p), c_int]
network_predict = lib.network_predict
network_predict.argtypes = [c_void_p, POINTER(c_float)]
reset_rnn = lib.reset_rnn
reset_rnn.argtypes = [c_void_p]
load_net = lib.load_network
load_net.argtypes = [c_char_p, c_char_p, c_int]
load_net.restype = c_void_p
do_nms_obj = lib.do_nms_obj
do_nms_obj.argtypes = [POINTER(DETECTION), c_int, c_int, c_float]
do_nms_sort = lib.do_nms_sort
do_nms_sort.argtypes = [POINTER(DETECTION), c_int, c_int, c_float]
free_image = lib.free_image
free_image.argtypes = [IMAGE]
letterbox_image = lib.letterbox_image
letterbox_image.argtypes = [IMAGE, c_int, c_int]
letterbox_image.restype = IMAGE
load_meta = lib.get_metadata
lib.get_metadata.argtypes = [c_char_p]
lib.get_metadata.restype = METADATA
load_image = lib.load_image_color
load_image.argtypes = [c_char_p, c_int, c_int]
load_image.restype = IMAGE
rgbgr_image = lib.rgbgr_image
rgbgr_image.argtypes = [IMAGE]
predict_image = lib.network_predict_image
predict_image.argtypes = [c_void_p, IMAGE]
predict_image.restype = POINTER(c_float)
def classify(net, meta, im):
out = predict_image(net, im)
res = []
for i in range(meta.classes):
res.append((meta.names[i], out[i]))
print(res)
res = sorted(res, key=lambda x: -x[1])
return res
def detect(net, meta, image, thresh=.6, hier_thresh=.7, nms=.45):
im = load_image(image, 0, 0)
num = c_int(0)
pnum = pointer(num)
predict_image(net, im)
dets = get_network_boxes(net, im.w, im.h, thresh, hier_thresh, None, 0, pnum)
num = pnum[0]
if (nms): do_nms_obj(dets, num, meta.classes, nms);
res = []
for j in range(num):
for i in range(meta.classes):
if dets[j].prob[i] > 0:
b = dets[j].bbox
res.append((meta.names[i], dets[j].prob[i], (b.x, b.y, b.w, b.h)))
res = sorted(res, key=lambda x: -x[1])
free_image(im)
free_detections(dets, num)
return res
def interface(net, meta, outpath):
rrr = detect(net, meta, outpath)
return rrr
if name == “main”:
if os.path.exists('//test/'):
shutil.rmtree('//test/')
os.mkdir('//test/')
path = "//tupian/"
filelist = os.listdir(path)
for image1 in filelist:
image2 = image1.split('.')[0]
adr = path + image1
image = cv2.imread(adr)
size = (1230, 2390)
image = cv2.resize(image, size, interpolation=cv2.INTER_AREA)
imgs = clip(image, 150, 800, 800)
for i in range(len(imgs)):
cv2.imwrite('/et/AXI3/test/' + str(image2) + str(i) + '.jpg',
imgs[i])
net = load_net("/ho/darknet/AXI3/yolov3-obj1.cfg", "/het/backup/yolov3-obj1_30000.weights", 0)
meta = load_meta("/h/darknet/AXI3/axi.data")
path='/knet//test'
path1='//dest'
if os.path.exists('/dest'):
shutil.rmtree('/dest')
os.mkdir('dest')
list=[]
for filename in os.listdir(path):
img = cv2.imread(path + "/" + filename)
outpath=path+'/'+filename
rrr=interface(net, meta, outpath)
for item in rrr:
cv2.rectangle(img,(int(item[2][0]-item[2][2]/2),int(item[2][1]-item[2][3]/2)),(int(item[2][0]+item[2][2]/2),int(item[2][1]+item[2][3]/2)),(0,255,0),2)
cv2.putText(img,str(item[0]),(int(item[2][0]-item[2][2]/2),int(item[2][1]-item[2][3]/2)+20),cv2.FONT_HERSHEY_SIMPLEX,1,[0,255,0],4)
if len(rrr)>0:
cv2.imwrite(path1 + "/" + filename,img)
list.append(filename.split('.')[0][:-1]+'.jpg')
print(len(list))
lists = np.unique(list)
print(len(lists))
