1.案例效果:
2.案例需求:
代码:
对结果进行标记-完整代码:
def tag_picture(self, images_data, outputs):
"""
显示预测结果到图片中
:return:
"""
# 获取每张图片预测结果的值
for i, img in enumerate(images_data):
# 解析输出结果,每张图片的标签,置信度和位置
pre_label = outputs[i][:, 0]
pre_conf = outputs[i][:, 1]
pre_xmin = outputs[i][:, 2]
pre_ymin = outputs[i][:, 3]
pre_xmax = outputs[i][:, 4]
pre_ymax = outputs[i][:, 5]
print("label: {}, probability: {}, pre_xmin: {}, pre_ymin: {}, pre_xmax: {}, pre_ymax: {}".format(pre_label, pre_conf, pre_xmin, pre_ymin, pre_xmax, pre_ymax))
# 由于检测出来的物体还是很多,所以进行显示过滤
top_indices = [i for i, conf in enumerate(pre_conf) if conf >= 0.6]
top_conf = pre_conf[top_indices]
top_label_indices = pre_label[top_indices].tolist()
top_xmin = pre_xmin[top_indices]
top_ymin = pre_ymin[top_indices]
top_xmax = pre_xmax[top_indices]
top_ymax = pre_ymax[top_indices]
# 画图
colors = plt.cm.hsv(np.linspace(0, 1, 21)).tolist()
plt.imshow(img / 255.)
currentAxis = plt.gca()
for i in range(top_conf.shape[0]):
xmin = int(round(top_xmin[i] * img.shape[1]))
ymin = int(round(top_ymin[i] * img.shape[0]))
xmax = int(round(top_xmax[i] * img.shape[1]))
ymax = int(round(top_ymax[i] * img.shape[0]))
# 获取该图片预测概率,名称,定义显示颜色
score = top_conf[i]
label = int(top_label_indices[i])
label_name = self.classes_name[label - 1]
display_txt = '{:0.2f}, {}'.format(score, label_name)
coords = (xmin, ymin), xmax - xmin + 1, ymax - ymin + 1
color = colors[label]
# 显示方框
currentAxis.add_patch(plt.Rectangle(*coords, fill=False, edgecolor=color, linewidth=2))
# 左上角显示概率以及名称
currentAxis.text(xmin, ymin, display_txt, bbox={'facecolor': color, 'alpha': 0.5})
plt.show()
完整代码:
from nets.ssd_net import SSD300
from utils.ssd_utils import BBoxUtility
from tensorflow.keras.preprocessing.image import load_img, img_to_array
import os
from scipy.misc import imread
from keras.applications.imagenet_utils import preprocess_input
import numpy as np
import matplotlib.pyplot as plt
class SSDTest(object):
def __init__(self):
# 定义识别类别
self.classes_name = ['Aeroplane', 'Bicycle', 'Bird', 'Boat', 'Bottle',
'Bus', 'Car', 'Cat', 'Chair', 'Cow', 'Diningtable',
'Dog', 'Horse', 'Motorbike', 'Person', 'Pottedplant',
'Sheep', 'Sofa', 'Train', 'Tvmonitor']
# 定义模型输入的参数 1 北京
self.classes_nums = len(self.classes_name) + 1
self.input_shape = (300, 300, 3)
def test(self):
"""
对于输入的图片进行预测物体位置
:return:
"""
# SSD300模型输入以及加载参数
model = SSD300(self.input_shape, num_classes=self.classes_nums)
model.load_weights("./ckpt/transfer_01-0.64.h5", by_name=True)
feature = []
images_data = []
# 读取多个本地路径测试图片, preprocessing_input以及保存图像像素值(显示需要)
for path in os.listdir("./images"):
img_path = os.path.join("./images/", path)
# 1. 输入到SSD网络中。数组
image = load_img(img_path, target_size=(self.input_shape[0], self.input_shape[1]))
image = img_to_array(image)
feature.append(image)
# 2。 读取图片二进制数据
images_data.append(imread(img_path))
# 模型预测结果,得到7308个priorbox
# 处理
inputs = preprocess_input(np.asarray(feature))
pred = model.predict(inputs)
# res形状 (2, 7308, 33) 2代表图片数量,7308代表每个图片预测的default boxes数量, 33:4(位置) + 21(预测概率) + 8(其他default boxes参数)
# 进行非最大抑制算放处理
bb = BBoxUtility(self.classes_nums)
res = bb.detection_out(pred)
# res (200, 6) (200, 6)
print(res[0].shape, res[1].shape)
# 还剩200 个候选框, 每个位置类别都是6
return res, images_data
def tag_picture(self, images_data, outputs):
"""
显示预测结果到图片中
:return:
"""
# 获取每张图片预测结果的值
for i, img in enumerate(images_data):
# 解析输出结果,每张图片的标签,置信度和位置
pre_label = outputs[i][:, 0]
pre_conf = outputs[i][:, 1]
pre_xmin = outputs[i][:, 2]
pre_ymin = outputs[i][:, 3]
pre_xmax = outputs[i][:, 4]
pre_ymax = outputs[i][:, 5]
print("label: {}, probability: {}, pre_xmin: {}, pre_ymin: {}, pre_xmax: {}, pre_ymax: {}".format(pre_label, pre_conf, pre_xmin, pre_ymin, pre_xmax, pre_ymax))
# 由于检测出来的物体还是很多,所以进行显示过滤
top_indices = [i for i, conf in enumerate(pre_conf) if conf >= 0.6]
top_conf = pre_conf[top_indices]
top_label_indices = pre_label[top_indices].tolist()
top_xmin = pre_xmin[top_indices]
top_ymin = pre_ymin[top_indices]
top_xmax = pre_xmax[top_indices]
top_ymax = pre_ymax[top_indices]
# 画图
# 定义21中颜色
colors = plt.cm.hsv(np.linspace(0, 1, 21)).tolist()
plt.imshow(img / 255.)
currentAxis = plt.gca()
for i in range(top_conf.shape[0]):
xmin = int(round(top_xmin[i] * img.shape[1]))
ymin = int(round(top_ymin[i] * img.shape[0]))
xmax = int(round(top_xmax[i] * img.shape[1]))
ymax = int(round(top_ymax[i] * img.shape[0]))
# 获取该图片预测概率,名称,定义显示颜色
score = top_conf[i]
label = int(top_label_indices[i])
label_name = self.classes_name[label - 1]
display_txt = '{:0.2f}, {}'.format(score, label_name)
coords = (xmin, ymin), xmax - xmin + 1, ymax - ymin + 1
color = colors[label]
# 显示方框
currentAxis.add_patch(plt.Rectangle(*coords, fill=False, edgecolor=color, linewidth=2))
# 左上角显示概率以及名称
currentAxis.text(xmin, ymin, display_txt, bbox={'facecolor': color, 'alpha': 0.5})
plt.show()
return None
if __name__ == '__main__':
ssd = SSDTest()
outputs, images_data = ssd.test()
# 显示图片
ssd.tag_picture(images_data, outputs)