update 121_125(DNN) and model file

Author: JimmyHHua

README.md

@@ -158,4 +158,9 @@ code_116 | [Decision Tree Algorithm](python/code_116)   | ✔️
 code_117 | [Image Mean-shift Segmentation](python/code_117)   | ✔️
 code_118 | [Grabcut-Image Segmentation](python/code_118)   | ✔️
 code_119 | [Grabcut-Background Change](python/code_119)   | ✏️
-code_120 | [Qrcode detect and decode](python/code_120)   | ✏️
+code_120 | [Qrcode detect and decode](python/code_120)   | ✏️
+code_121 | [DNN- Read the information of each layer of the model](python/code_121)   | ✔️
+code_122 | [DNN- Realize image classification](python/code_122)   | ✔️
+code_120 | [DNN- Model runs to set the target device and compute the background](python/code_123)   | ✔️
+code_120 | [DNN- SSD Single Image Detection](python/code_124)   | ✔️
+code_120 | [DNN- SSD Real-time Video Detection](python/code_125)   | ✔️

README_CN.md

@@ -157,4 +157,9 @@ code_116 | [决策树算法](python/code_116)   | ✔️
 code_117 | [图像均值漂移分割](python/code_117)   | ✔️
 code_118 | [Grabcut-图像分割](python/code_118)   | ✔️
 code_119 | [Grabcut-背景替换](python/code_119)   | ✏️
-code_120 | [二维码检测识别](python/code_120)   | ✔️
+code_120 | [二维码检测识别](python/code_120)   | ✔️
+code_121 | [DNN- 读取模型各层信息](python/code_121)   | ✔️
+code_122 | [DNN- DNN实现图像分类](python/code_122)   | ✔️
+code_120 | [DNN- 模型运行设置目标设备与计算后台](python/code_123)   | ✔️
+code_120 | [DNN- SSD单张图片检测](python/code_124)   | ✔️
+code_120 | [DNN- SSD实时视频检测](python/code_125)   | ✔️

python/code_121/README.md

@@ -0,0 +1,71 @@
+## 读取DNN模型各层信息
+
+### 概述
+ ✔️ OpenCV的DNN模块支持下面框架的预训练模型的前馈网络(预测图)使用：
+- Caffe
+- Tensorflow
+- Torch
+- DLDT
+- Darknet
+
+同时还支持自定义层解析、非最大抑制操作、获取各层的信息等。
+OpenCV加载模型的通用API为:
+```
+cv2.dnn.readNet(model,  # 模型
+            	config = "", 
+            	framework = "" )
+```
+其中：
+- model二进制训练好的网络权重文件，可能来自支持的网络框架，扩展名为如下：
+
+    - *.caffemodel (Caffe,http://caffe.berkeleyvision.org/)
+    - *.pb (TensorFlow, https://www.tensorflow.org/)
+    - *.t7 | *.net (Torch, http://torch.ch/)
+    - *.weights (Darknet, https://pjreddie.com/darknet/)
+    - *.bin (DLDT, https://software.intel.com/openvino-toolkit)
+
+- config针对模型二进制的描述文件，不同的框架配置文件有不同扩展名：
+
+    - *.prototxt (Caffe, http://caffe.berkeleyvision.org/)
+    - *.pbtxt (TensorFlow, https://www.tensorflow.org/)
+    - *.cfg (Darknet, https://pjreddie.com/darknet/)
+    - *.xml (DLDT, https://software.intel.com/openvino-toolkit)
+
+- framework显示声明参数，说明模型使用哪个框架训练出来的。
+
+### 代码
+```python
+import cv2
+import numpy as np
+
+bin_model = "bvlc_googlenet.caffemodel"
+protxt = "bvlc_googlenet.prototxt"
+
+# load CNN model
+net = cv2.dnn.readNet(bin_model, protxt)
+
+# 获取各层信息
+layer_names = net.getLayerNames()
+
+for name in layer_names:
+    id = net.getLayerId(name)
+    layer = net.getLayer(id)
+    print("layer id : %d, type : %s, name: %s"%(id, layer.type, layer.name))
+
+print("successfully")
+```
+>输出
+```
+layer id : 1, type : Convolution, name: conv1/7x7_s2
+layer id : 2, type : ReLU, name: conv1/relu_7x7
+layer id : 3, type : Pooling, name: pool1/3x3_s2
+layer id : 4, type : LRN, name: pool1/norm1
+layer id : 5, type : Convolution, name: conv2/3x3_reduce
+layer id : 6, type : ReLU, name: conv2/relu_3x3_reduce
+layer id : 7, type : Convolution, name: conv2/3x3
+layer id : 8, type : ReLU, name: conv2/relu_3x3
+layer id : 9, type : LRN, name: conv2/norm2
+layer id : 10, type : Pooling, name: pool2/3x3_s2
+...
+successfully
+```

python/code_121/guinea_pig.jpg

@@ -0,0 +1,25 @@
+import cv2
+import numpy as np
+
+bin_model = "../model/google/bvlc_googlenet.caffemodel"
+protxt = "../model/google/bvlc_googlenet.prototxt"
+
+# Load names of classes
+# classes = None
+# with open("classification_classes_ILSVRC2012.txt", 'rt') as f:
+#     classes = f.read().rstrip('\n').split('\n')
+
+# print(classes)
+
+# load CNN model
+net = cv2.dnn.readNet(bin_model, protxt)
+
+# 获取各层信息
+layer_names = net.getLayerNames()
+
+for name in layer_names:
+    id = net.getLayerId(name)
+    layer = net.getLayer(id)
+    print("layer id : %d, type : %s, name: %s"%(id, layer.type, layer.name))
+
+print("successfully loaded model...")

python/code_121/opencv_121.py

@@ -0,0 +1,83 @@
+## DNN实现图像分类
+
+### 概述
+
+ ✔️ 我们使用ImageNet数据集支持1000分类的GoogleNet网络模型， 结合opencv实现图像分类标签预测。
+
+>label标签是在一个单独的文本文件中读取。
+
+读取模型的API：
+```python
+cv2.dnn.readNetFromCaffe(prototxt,
+                         caffeModel = String())
+```
+其中：
+- prototxt表示模型的配置文件
+- caffeModel表示模型的权重二进制文件
+
+使用模型实现预测的时候，需要读取图像作为输入，网络模型支持的输入数据是四维的输入，所以要把读取到的Mat对象转换为四维张量，OpenCV的提供的API为如下：
+
+```python
+cv2.dnn。blobFromImage(
+                        image,
+                        scalefactor = 1.0,
+                        size = Size(),
+                        mean = Scalar(),
+                        swapRB = false,
+                        crop = false,
+                        ddepth = CV_32F 
+)
+```
+其中：
+- image输入图像
+- scalefactor 默认1.0
+- size表示网络接受的数据大小
+- mean表示训练时数据集的均值
+- swapRB 是否互换Red与Blur通道
+- crop剪切
+- ddepth 数据类型
+
+### 代码
+```python
+import cv2
+import numpy as np
+
+bin_model = "bvlc_googlenet.caffemodel"
+protxt = "bvlc_googlenet.prototxt"
+
+# 加载类别
+classes = None
+with open("classification_classes_ILSVRC2012.txt", 'rt') as f:
+    classes = f.read().rstrip('\n').split('\n')
+
+# 加载模型
+net = cv2.dnn.readNetFromCaffe(protxt, bin_model)
+
+# 读取输入数据
+image = cv2.imread("guinea_pig.jpg")
+blob = cv2.dnn.blobFromImage(image, 1.0, (224, 224), (104, 117,123), False, crop=False)
+result = np.copy(image)
+
+# 运行模型
+net.setInput(blob)
+out = net.forward()
+
+# 获取最高分的类别
+out = out.flatten()
+classId = np.argmax(out)
+confidence = out[classId]
+
+# 输出运行时间
+t, _ = net.getPerfProfile() #返回值是网络执行推断的时间
+label = 'cost time: %.2f ms' % (t * 1000.0 / cv2.getTickFrequency())
+cv2.putText(result, label, (0, 20), cv2.FONT_HERSHEY_SIMPLEX, 0.8, (255, 255, 0), 2)
+
+# 显示结果
+label = '%s: %.4f' % (classes[classId] if classes else 'Class #%d' % classId, confidence)
+cv2.putText(result, label, (0, 60), cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 0, 255), 2)
+
+show_img = np.hstack((image, result))
+```
+>输出豚鼠的预测结果
+
+![result]('./result.jpg')

python/code_122/README.md

@@ -0,0 +1,43 @@
+import cv2
+import numpy as np
+
+bin_model = "../model/google/bvlc_googlenet.caffemodel"
+protxt = "../model/google/bvlc_googlenet.prototxt"
+
+# Load names of classes
+classes = None
+with open("../model/google/classification_classes_ILSVRC2012.txt", 'rt') as f:
+    classes = f.read().rstrip('\n').split('\n')
+
+# load CNN model
+net = cv2.dnn.readNetFromCaffe(protxt, bin_model)
+
+# read input data
+image = cv2.imread("guinea_pig.jpg")
+blob = cv2.dnn.blobFromImage(image, 1.0, (224, 224), (104, 117,123), False, crop=False)
+result = np.copy(image)
+cv2.imshow("input", image)
+
+# Run a model
+net.setInput(blob)
+out = net.forward()
+
+# Get a class with a highest score.
+out = out.flatten()
+classId = np.argmax(out)
+confidence = out[classId]
+
+# Put efficiency information.
+t, _ = net.getPerfProfile()
+label = 'cost time: %.2f ms' % (t * 1000.0 / cv2.getTickFrequency())
+cv2.putText(result, label, (0, 20), cv2.FONT_HERSHEY_SIMPLEX, 0.8, (255, 255, 0), 2)
+
+# Print predicted class.
+label = '%s: %.4f' % (classes[classId] if classes else 'Class #%d' % classId, confidence)
+cv2.putText(result, label, (0, 60), cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 0, 255), 2)
+
+show_img = np.hstack((image, result))
+cv2.namedWindow('demo', cv2.WINDOW_NORMAL)
+cv2.imshow("demo", show_img)
+cv2.waitKey(0)
+cv2.destroyAllWindows()

python/code_122/dog.jpg

@@ -0,0 +1,18 @@
+## OpenCV DNN 为模型运行设置目标设备与计算后台
+
+OpenCV中加载网络模型之后，可以设置计算后台与计算目标设备，OpenCV DNN模块支持这两个设置的相关API如下：
+cv::dnn::Net::setPreferableBackend(
+	int backendId
+)
+backendId 表示后台计算id，
+-	DNN_BACKEND_DEFAULT (DNN_BACKEND_INFERENCE_ENGINE)表示默认使用intel的预测推断库(需要下载安装Intel® OpenVINO™ toolkit， 然后重新编译OpenCV源码，在CMake时候enable该选项方可)， 可加速计算！
+-	DNN_BACKEND_OPENCV 一般情况都是使用opencv dnn作为后台计算，
+
+void cv::dnn::Net::setPreferableTarget(
+	int targetId
+)
+常见的目标设备id如下：
+-	DNN_TARGET_CPU其中表示使用CPU计算，默认是的
+-	DNN_TARGET_OPENCL 表示使用OpenCL加速，一般情况速度都很扯
+-	DNN_TARGET_OPENCL_FP16 可以尝试
+-	DNN_TARGET_MYRIAD 树莓派上的

python/code_122/guinea_pig.jpg

@@ -0,0 +1,45 @@
+import cv2
+import numpy as np
+
+bin_model = "../model/google/bvlc_googlenet.caffemodel"
+protxt = "../model/google/bvlc_googlenet.prototxt"
+
+# Load names of classes
+classes = None
+with open("../model/google/classification_classes_ILSVRC2012.txt", 'rt') as f:
+    classes = f.read().rstrip('\n').split('\n')
+
+# load CNN model
+net = cv2.dnn.readNetFromCaffe(protxt, bin_model)
+net.setPreferableBackend(cv2.dnn.DNN_BACKEND_OPENCV)
+net.setPreferableTarget(cv2.dnn.DNN_TARGET_CPU)
+
+# read input data
+image = cv2.imread("guinea_pig.jpg")
+blob = cv2.dnn.blobFromImage(image, 1.0, (224, 224), (104, 117,123), False, crop=False)
+result = np.copy(image)
+cv2.imshow("input", image)
+
+# Run a model
+net.setInput(blob)
+out = net.forward()
+
+# Get a class with a highest score.
+out = out.flatten()
+classId = np.argmax(out)
+confidence = out[classId]
+
+# Put efficiency information.
+t, _ = net.getPerfProfile()
+label = 'cost time: %.2f ms' % (t * 1000.0 / cv2.getTickFrequency())
+cv2.putText(result, label, (0, 20), cv2.FONT_HERSHEY_SIMPLEX, 0.8, (255, 255, 0), 2)
+
+# Print predicted class.
+label = '%s: %.4f' % (classes[classId] if classes else 'Class #%d' % classId, confidence)
+cv2.putText(result, label, (0, 60), cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 0, 255), 2)
+
+show_img = np.hstack((image, result))
+cv2.namedWindow('demo', cv2.WINDOW_NORMAL)
+cv2.imshow("demo", show_img)
+cv2.waitKey(0)
+cv2.destroyAllWindows()

python/code_122/opencv_122.py

@@ -0,0 +1,56 @@
+## SSD实现目标检测
+
+✔️ OpenCV的DNN模块支持常见得对象检测模型SSD， 以及Mobile Net-SSD。
+
+✔️ 这里我们用基于Caffe训练好的mobile-net SSD来测试目标检测。
+
+### 单张图片检测
+
+```python
+import cv2
+
+model_bin = "../model/ssd/MobileNetSSD_deploy.caffemodel";
+config_text = "../model/ssd/MobileNetSSD_deploy.prototxt";
+objName = ["background",
+			"aeroplane", "bicycle", "bird", "boat",
+			"bottle", "bus", "car", "cat", "chair",
+			"cow", "diningtable", "dog", "horse",
+			"motorbike", "person", "pottedplant",
+			"sheep", "sofa", "train", "tvmonitor"];
+
+# 加载模型
+net = cv2.dnn.readNetFromCaffe(config_text, model_bin)
+image = cv2.imread("dog.jpg")
+h = image.shape[0]
+w = image.shape[1]
+
+# 获得所有层名称与索引
+layerNames = net.getLayerNames()
+lastLayerId = net.getLayerId(layerNames[-1])
+lastLayer = net.getLayer(lastLayerId)
+print(lastLayer.type)
+
+# 加载图片检测
+blobImage = cv2.dnn.blobFromImage(image, 0.007843, (300, 300), (127.5, 127.5, 127.5), True, False);
+net.setInput(blobImage)
+Out = net.forward()
+print(Out.shape)
+for detection in Out[0,0,:,:]:
+    score = float(detection[2])
+    objIndex = int(detection[1])
+    if score > 0.5:
+        left = detection[3]*w
+        top = detection[4]*h
+        right = detection[5]*w
+        bottom = detection[6]*h
+
+        # 绘制
+        cv2.rectangle(image, (int(left), int(top)), (int(right), int(bottom)), (255, 0, 0), thickness=2)
+        cv2.putText(image, "score:%.2f, %s"%(score, objName[objIndex]),
+                (int(left) - 10, int(top) - 5), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2, 8);
+
+```
+>输出
+
+![dog]('./dog.jpg)
+![Pedestrian]('./result.png)