Wykrywanie_zwierzat/Find_animal.py

161 lines
5.0 KiB
Python
Raw Normal View History

2020-01-10 22:07:25 +01:00
import cv2 as cv
import argparse
import numpy as np
import os.path
import sys
import random
# Inicjalizacja parametrów
confThreshold = 0.5
maskThreshold = 0.3
args = parser.parse_args()
# Rysuje obrawmowanie zwierzęcia, koloruje i zaznacza maską
def drawBox(frame, classId, conf, left, top, right, bottom, classMask):
# obramowanie.
cv.rectangle(frame, (left, top), (right, bottom), (255, 178, 50), 3)
# etykieta obiektu
label = '%.2f' % conf
if classes:
assert(classId < len(classes))
label = '%s:%s' % (classes[classId], label)
# wyświetla etykietę
labelSize, baseLine = cv.getTextSize(label, cv.FONT_HERSHEY_SIMPLEX, 0.5, 1)
top = max(top, labelSize[1])
cv.rectangle(frame, (left, top - round(1.5*labelSize[1])), (left + round(1.5*labelSize[0]), top + baseLine), (255, 255, 255), cv.FILLED)
cv.putText(frame, label, (left, top), cv.FONT_HERSHEY_SIMPLEX, 0.75, (0,0,0), 1)
# zmiana rozmiaru maski i nałożenie na obiekt
classMask = cv.resize(classMask, (right - left + 1, bottom - top + 1))
mask = (classMask > maskThreshold)
roi = frame[top:bottom+1, left:right+1][mask]
colorIndex = random.randint(0, len(colors)-1)
color = colors[colorIndex]
frame[top:bottom+1, left:right+1][mask] = ([0.3*color[0], 0.3*color[1], 0.3*color[2]] + 0.7 * roi).astype(np.uint8)
# rysuje kontury na obrazie
mask = mask.astype(np.uint8)
im2, contours, hierarchy = cv.findContours(mask,cv.RETR_TREE,cv.CHAIN_APPROX_SIMPLE)
cv.drawContours(frame[top:bottom+1, left:right+1], contours, -1, color, 3, cv.LINE_8, hierarchy, 100)
# dla każdej ramki maskuje obraz
def postprocess(boxes, masks):
# N - liczba znalezionych obramowań
# C - liczba klas
# H,W- wysokość i szerokość
numClasses = masks.shape[1]
numDetections = boxes.shape[2]
frameH = frame.shape[0]
frameW = frame.shape[1]
for i in range(numDetections):
box = boxes[0, 0, i]
mask = masks[i]
score = box[2]
if score > confThreshold:
classId = int(box[1])
# zaznacza ramkę
left = int(frameW * box[3])
top = int(frameH * box[4])
right = int(frameW * box[5])
bottom = int(frameH * box[6])
left = max(0, min(left, frameW - 1))
top = max(0, min(top, frameH - 1))
right = max(0, min(right, frameW - 1))
bottom = max(0, min(bottom, frameH - 1))
# aktywacja maski
classMask = mask[classId]
# rysuje wszystko na obrazie
drawBox(frame, classId, score, left, top, right, bottom, classMask)
# załaduj nazwy
2020-01-11 13:15:36 +01:00
classesFile = "labels.names";
2020-01-10 22:07:25 +01:00
classes = None
with open(classesFile, 'rt') as f:
classes = f.read().rstrip('\n').split('\n')
# uruchomienie grafu tekstowego i modelu wagi
textGraph = "./mask.pbtxt";
modelWeights = "./mask/frozen_inference_graph.pb";
# Załadowanie z sieci
2020-01-11 13:13:00 +01:00
#net = cv.dnn.readNetFromTensorflow(modelWeights, textGraph);
#net.setPreferableBackend(cv.dnn.DNN_BACKEND_OPENCV)
#net.setPreferableTarget(cv.dnn.DNN_TARGET_CPU)
2020-01-10 22:07:25 +01:00
# załadowanie klas
colorsFile = "colors.txt";
with open(colorsFile, 'rt') as f:
colorsStr = f.read().rstrip('\n').split('\n')
colors = [] #[0,0,0]
for i in range(len(colorsStr)):
rgb = colorsStr[i].split(' ')
color = np.array([float(rgb[0]), float(rgb[1]), float(rgb[2])])
colors.append(color)
winName = 'Animal finding program'
cv.namedWindow(winName, cv.WINDOW_NORMAL)
outputFile = "found_animal.avi"
if (args.image):
# Otrwiera pllik z obrazem
if not os.path.isfile(args.image):
print("Input image file ", args.image, " doesn't exist")
sys.exit(1)
cap = cv.VideoCapture(args.image)
2020-01-11 13:15:36 +01:00
outputFile = args.image[:-4]+'found_animal.jpg'
2020-01-10 22:07:25 +01:00
elif (args.video):
# Otwiera plik video
if not os.path.isfile(args.video):
print("Input video file ", args.video, " doesn't exist")
sys.exit(1)
cap = cv.VideoCapture(args.video)
2020-01-11 13:15:36 +01:00
outputFile = args.video[:-4]+'found_animal.avi'
2020-01-10 22:07:25 +01:00
# inicjalizacja zapisu video
if (not args.image):
vid_writer = cv.VideoWriter(outputFile, cv.VideoWriter_fourcc('M','J','P','G'), 28, (round(cap.get(cv.CAP_PROP_FRAME_WIDTH)),round(cap.get(cv.CAP_PROP_FRAME_HEIGHT))))
while cv.waitKey(1) < 0:
hasFrame, frame = cap.read()
# zamyka program, gdy skończy się film
if not hasFrame:
2020-01-11 13:15:36 +01:00
print("Processing Done.")
print("Output file saved ad: ", outputFile)
2020-01-10 22:07:25 +01:00
cv.waitKey(3000)
break
# Utworzenie boxa
blob = cv.dnn.blobFromImage(frame, swapRB=True, crop=False)
net.setInput(blob)
# Uruchomienie przekazanie, aby uzyskać dane wyjściowe
boxes, masks = net.forward(['detection_out_final', 'detection_masks'])
# Wyrzuca obramowanie i maskę dla każdego wykrytego obiektu
postprocess(boxes, masks)
# Zapisuje obraz z obramowaniami
if (args.image):
cv.imwrite(outputFile, frame.astype(np.uint8));
else:
vid_writer.write(frame.astype(np.uint8))
cv.imshow(winName, frame)