def get_features(features,labels):
pos_imgs,pos_labels = read_pos_samples('F:/za/Model/pedestrain/INRIAPerson/pos_64x128')#==============================================
computeHog(pos_imgs,features)
[labels.append(1) for _ in range(len(pos_imgs))]
neg_imgs,neg_labels = read_neg_samples('F:/za/Model/pedestrain/INRIAPerson/neg_64x128')
computeHog(neg_imgs,features)
[labels.append(-1) for _ in range(len(neg_imgs))]
#print('feature_shape = ',np.shape(features))
return features,labels
def get_features(features, labels):
print('compute pos features...')
pos_imgs, pos_labels = read_pos_samples('E:/pypj/pos')
hog_compute(pos_imgs, features)
[labels.append(1) for _ in range(len(pos_imgs))]
print('compute negs features...')
neg_imgs, neg_labels = read_neg_samples('E:/pypj/negs')
hog_compute(neg_imgs, features)
[labels.append(-1) for _ in range(len(neg_imgs))]
print('features get!')
def get_features(features, labels):
print('get pos hog...')
pos_imgs, pos_labels = read_pos_samples(
'C:/Users/fyyk_whua/Desktop/python demo/MachineLearn/hog+svm/Img/pos_250x250/'
)
print('compute pos hog...')
computeHog(pos_imgs, features)
print('append pos labels...')
[labels.append(1) for _ in range(len(pos_imgs))]
print('get neg hog...')
neg_imgs, neg_labels = read_neg_samples(
'C:/Users/fyyk_whua/Desktop/python demo/MachineLearn/hog+svm/Img/neg_250x250/'
)
print('compute neg hog...')
computeHog(neg_imgs, features)
print('append neg labels...')
[labels.append(-1) for _ in range(len(neg_imgs))]
return features, labels
import cv2
import numpy as np
from get_data import read_pos_samples, read_neg_samples
from svm_train import svm_config, svm_train, svm_save, svm_load, get_svm_detector
from get_hog import hog_compute, hog_config, get_features
from tqdm import tqdm
if __name__ == '__main__':
features = []
labels = []
pos_imgs, pos_labels = read_pos_samples('E:/pypj/res')
get_features(features, labels)
scores = []
result = []
print('start validate')
pbar = tqdm(total=100)
step = (-5, -4, -3, -2, -1, 0, 1, 2, 3, 4, 5)
for i in step:
C = 2**i
for j in step:
G = 2**j
svm = svm_config(C, G)
svm_train(svm, features, labels)
hog = hog_config()
hog.setSVMDetector(get_svm_detector(svm))
for pos_img in pos_imgs:
rects, score = hog.detect(pos_img)
if len(score) == 0:
score = np.array([0])
scores.append(score.mean())
result.append([C, G, np.array(scores).mean()])