add for hat model

examples/python/hatdet/README

+第一：环境
+#运行该项目，需要安装python版本opencv4.4.0以上版本;
+如需卸载sudo pip3 uninstall opencv-python
+安装最新版本：sudo pip3 install opencv-python
+查看版本>>python3
+>>>import cv2
+>>>cv2.__version__
+>>>quit() 
+
+第二：运行
+python3 test_darknet.py --image 6.jpg
+python3 test_darknet.py --video 0904.mp4 
+
+yolo.getdetresults(frame,False)//接口设置False为传递结果参数，设置True为测试
+
+rtsp://admin:admin12345@192.168.77.19:554/h264/ch1/main/av_stream

examples/python/hatdet/darknet.py

+import cv2
+import numpy as np
+import os.path
+
+class DarknetYolo:
+    def __init__(self,configPath, weightPath, metaPath, thresh):
+        assert 0 < thresh < 1, "Threshold should be a float between zero and one (non-inclusive)"
+        if not os.path.exists(configPath):
+            raise ValueError("Invalid config path `" + os.path.abspath(configPath) + "`")
+        if not os.path.exists(weightPath):
+            raise ValueError("Invalid weight path `" + os.path.abspath(weightPath) + "`")
+        if not os.path.exists(metaPath):
+            raise ValueError("Invalid data file path `" + os.path.abspath(metaPath) + "`")
+        #if netMain is None:
+        netMain = cv2.dnn.readNetFromDarknet(configPath, weightPath)  
+        netMain.setPreferableBackend(cv2.dnn.DNN_BACKEND_OPENCV)
+        netMain.setPreferableTarget(cv2.dnn.DNN_TARGET_CPU)
+        #if metaMain is None:
+        metaMain = open(metaPath).read().strip().split("\n")
+        
+        np.random.seed(666)
+        COLORS = np.random.randint(0, 255, size=(len(metaMain), 3), dtype="uint8")
+        
+        self.netMain=netMain
+        self.metaMain=metaMain
+        self.thresh=thresh
+        self.COLORS=COLORS
+
+
+    def getdetresults(self, frame, hasdraw):
+        netMain, metaMain, thresh, COLORS = self.netMain, self.metaMain, self.thresh, self.COLORS
+        
+        # Create a 4D blob from a frame.
+        blob = cv2.dnn.blobFromImage(frame, 1/255, (416, 416), swapRB=True, crop=False)
+
+        # Sets the input to the network
+        netMain.setInput(blob)
+
+        # Runs the forward pass to get output of the output layers
+        layer = netMain.getUnconnectedOutLayersNames()
+        layerOutputs = netMain.forward(layer)
+
+        (H, W) = frame.shape[:2]
+        # Scan through all the bounding boxes output from the network and keep only the
+        # ones with high confidence scores. 
+        # Assign the box's class label as the class with the highest score.
+        boxes = []
+        confidences = []
+        classIDs = []
+
+        for output in layerOutputs:
+            for detection in output:
+                scores = detection[5:]
+                classID = np.argmax(scores)
+                confidence = scores[classID]
+                if confidence > thresh:
+                    box = detection[0:4] * np.array([W, H, W, H])
+                    (centerX, centerY, width, height) = box.astype("int")
+                    x = int(centerX - (width / 2))
+                    y = int(centerY - (height / 2))
+                    boxes.append([x, y, int(width), int(height)])
+                    confidences.append(float(confidence))
+                    classIDs.append(classID)
+        # Perform non maximum suppression to eliminate redundant overlapping boxes with
+        # lower confidences.
+        idxs = cv2.dnn.NMSBoxes(boxes, confidences, thresh, 0.3)
+        if len(idxs) > 0:
+            if not hasdraw: 
+                return idxs, boxes, confidences, classIDs, metaMain;            
+            for i in idxs.flatten():
+                # get x,y,w,h
+                (x, y) = (boxes[i][0], boxes[i][1])
+                (w, h) = (boxes[i][2], boxes[i][3])
+                # draw rectangle
+                color = [int(c) for c in COLORS[classIDs[i]]]
+                cv2.rectangle(frame, (x, y), (x + w, y + h), color, 1, lineType=cv2.LINE_AA)
+                text = "{}: {:.4f}".format(metaMain[classIDs[i]], confidences[i])
+                cv2.putText(frame, text, (x, y - 5), cv2.FONT_HERSHEY_SIMPLEX,
+                    0.5, color, 1, lineType=cv2.LINE_AA)
+                cv2.imshow("Tag", frame)
+                #cv2.waitKey(0)
+        else:
+            print("Cann't detect objects,please check image!")

examples/python/hatdet/hat.names

+hat
+person

examples/python/hatdet/test_darknet.py

+import cv2 as cv
+import os.path
+import darknet as interface
+import argparse
+
+
+configPath='hat-obj-test.cfg'
+weightPath='hat-obj_last.weights'
+metaPath='hat.names'
+thresh =0.9
+
+yolo=interface.DarknetYolo(configPath, weightPath, metaPath, thresh)
+
+parser = argparse.ArgumentParser(description='Object Detection using YOLO in OPENCV')
+parser.add_argument('--image', help='Path to image file.')
+parser.add_argument('--video', help='Path to video file.')
+parser.add_argument('--rtsp', help='Path to video file.')
+args = parser.parse_args()
+
+#outputFile = "yolo_out_py.avi"
+if (args.image):
+    # Open the image file
+    if not os.path.isfile(args.image):
+        print("Input image file ", args.image, " doesn't exist")
+        sys.exit(1)
+    cap = cv.VideoCapture(args.image)
+    #outputFile = args.image[:-4]+'_yolo_out_py.jpg'
+elif (args.video):
+    # Open the video file
+    if not os.path.isfile(args.video):
+        print("Input video file ", args.video, " doesn't exist")
+        sys.exit(1)
+    cap = cv.VideoCapture(args.video)
+    #outputFile = args.video[:-4]+'_yolo_out_py.avi'
+elif (args.rtsp):
+    # Open the video file
+    if not os.path.isfile(args.rtsp):
+        print("Input rtsp file ", args.rtsp, " doesn't exist")
+        sys.exit(1)
+    cap = cv.VideoCapture(args.rtsp)
+    #outputFile = args.video[:-4]+'_yolo_out_py.avi'
+else:
+    # Webcam input
+    cap = cv.VideoCapture(0)
+
+while cv.waitKey(1) < 0:
+    
+    # get frame from the video
+    hasFrame, frame = cap.read()
+    
+    # Stop the program if reached end of video
+    if not hasFrame:
+        print("Done processing !!!")
+        #print("Output file is stored as ", outputFile)
+        cv.waitKey(0)
+        # Release device
+        cap.release()
+        break
+
+    idxs, boxes, confidences, classIDs, metaMain = yolo.getdetresults(frame,False)
+    for i in idxs.flatten():
+        (x, y) = (boxes[i][0], boxes[i][1])
+        (w, h) = (boxes[i][2], boxes[i][3])
+        cv.rectangle(frame, (x, y), (x + w, y + h), (255, 255, 255), 1, lineType=cv.LINE_AA)
+        text = "{}: {:.4f}".format(metaMain[classIDs[i]], confidences[i])
+        cv.putText(frame, text, (x, y - 5), cv.FONT_HERSHEY_SIMPLEX,
+                    0.5, (255, 255, 255), 1, lineType=cv.LINE_AA)
+        cv.imshow("Tag", frame)
+