def image_callback(self, data):

        global theta
        global beta
        global tx
        global ty

        try:
            # Convert ROS image to OpenCV image
            raw_image = self.bridge.imgmsg_to_cv2(data, "bgr8")
        except CvBridgeError as e:
            print(e)

        # Flip the image 180 degrees
        cv_image = cv2.flip(raw_image, -1)

        # Draw a black line on the image
        cv2.line(cv_image, (0,50), (640,50), (0,0,0), 5)

        # cv_image is normal color image
        blob_image_center = blob_search(cv_image, self.detector)

        # Given world coordinate (xw, yw)
        xw = 0.2875
        yw = 0.1125

        # Only two blob center are found on the image
        if(len(blob_image_center) == 2):

            x1 = int(blob_image_center[0].split()[0])
            y1 = int(blob_image_center[0].split()[1])
            x2 = int(blob_image_center[1].split()[0])
            y2 = int(blob_image_center[1].split()[1])

            print("Blob Center 1: ({0}, {1}) and Blob Center 2: ({2}, {3})".format(x1, y1, x2, y2))

            ################################# Your Code Start Here ################################# 

            # Calculate beta, tx and ty, given x1, y1, x2, y2
            theta = 0
            beta = abs(x1-x2)/0.1
            if(x1 < x2):
            	c = x1
            	r = y1
            else:
            	c = x2
            	r = y2
            tx = (r-Or)/beta - xw
            ty = (c-Oc)/beta - yw

            ################################## Your Code End Here ################################## 

            print("theta = {0}\nbeta = {1}\ntx = {2}\nty = {3}\n".format(theta, beta, tx, ty))

        else:
            print("No Blob found! ")
    def image_callback(self, data):

        global theta
        global beta
        global tx
        global ty

        try:
            # Convert ROS image to OpenCV image
            raw_image = self.bridge.imgmsg_to_cv2(data, "bgr8")
        except CvBridgeError as e:
            print(e)

        # Flip the image 180 degrees
        cv_image = cv2.flip(raw_image, -1)

        # Draw a black line on the image
        cv2.line(cv_image, (0, 50), (640, 50), (0, 0, 0), 5)

        # cv_image is normal color image
        blob_image_center = blob_search(cv_image, self.detector)

        # Given world coordinate (xw, yw)
        xw = 0.2875
        yw = 0.1125

        # Only two blob center are found on the image
        if (len(blob_image_center) == 2):

            x1 = int(blob_image_center[0].split()[1])
            y1 = int(blob_image_center[0].split()[0])
            x2 = int(blob_image_center[1].split()[1])
            y2 = int(blob_image_center[1].split()[0])

            print("Blob Center 1: ({0}, {1}) and Blob Center 2: ({2}, {3})".
                  format(x1, y1, x2, y2))
            Or = 480 / 2
            Oc = 640 / 2
            beta = np.sqrt((x2 - x1)**2 + (y2 - y1)**2) / 0.1

            if (y1 < y2):
                tx = -xw + (x1 - Or) / beta
                ty = -yw + (y1 - Oc) / beta

            elif (y2 < y1):
                tx = -xw + (x2 - Or) / beta
                ty = -yw + (y2 - Oc) / beta

            ################################## Your Code End Here ##################################

            print("theta = {0}\nbeta = {1}\ntx = {2}\nty = {3}\n".format(
                theta, beta, tx, ty))

        else:
            print("No Blob found! ")
Example #3
0
    def image_callback(self, data):

        global theta
        global beta
        global tx
        global ty

        try:
            # Convert ROS image to OpenCV image
            raw_image = self.bridge.imgmsg_to_cv2(data, "bgr8")
        except CvBridgeError as e:
            print(e)

        # Flip the image 180 degrees
        cv_image = cv2.flip(raw_image, -1)

        # Draw a black line on the image
        cv2.line(cv_image, (0, 50), (640, 50), (0, 0, 0), 5)

        # cv_image is normal color image
        blob_image_center = blob_search(cv_image, self.detector)

        # Given world coordinate (xw, yw)
        xw = 0.2875
        yw = 0.1125

        # Only two blob center are found on the image
        if (len(blob_image_center) < 2):
            print("Not enough blobs found. " + str(len(blob_image_center)) +
                  " found.")

        elif (len(blob_image_center) == 2):

            c1 = int(blob_image_center[0].split()[0])
            r1 = int(blob_image_center[0].split()[1])
            c2 = int(blob_image_center[1].split()[0])
            r2 = int(blob_image_center[1].split()[1])

            print("Blob Center 1: ({0}, {1}) and Blob Center 2: ({2}, {3})".
                  format(c1, r1, c2, r2))

            ################################# Your Code Start Here #################################

            # Calculate beta, tx and ty, given c1, r1, c2, r2
            beta = 73 / 0.1

            ################################## Your Code End Here ##################################

            print("theta = {0}\nbeta = {1}\ntx = {2}\nty = {3}\n".format(
                theta, beta, tx, ty))

        else:
            print("To many points returned " + str(len(blob_image_center)) +
                  " found.")
Example #4
0
    def image_callback(self, data):

        global theta
        global beta
        global tx
        global ty

        try:
            # Convert ROS image to OpenCV image
            raw_image = self.bridge.imgmsg_to_cv2(data, "bgr8")
        except CvBridgeError as e:
            print(e)

        # Flip the image 180 degrees
        cv_image = cv2.flip(raw_image, -1)

        # Draw a black line on the image
        cv2.line(cv_image, (0, 50), (640, 50), (0, 0, 0), 5)

        # cv_image is normal color image
        blob_image_center = blob_search(cv_image, self.detector)

        if (len(blob_image_center) == 0):
            print("No blob found!")
            self.coord_pub.publish("")
        else:
            c = int(blob_image_center[0].split()[0])
            r = int(blob_image_center[0].split()[1])
            # print("Blob found! ({0}, {1})".format(c, r))

            ################################ Your Code Start Here ################################

            # Given theta, beta, tx, ty, calculate the world coordinate of c,r namely xw, yw
            theta = 0.0
            beta = 730.0
            tx = -0.282020547945
            ty = -0.150856164384

            position_c = np.array([[(r - 249) / beta], [(c - 369) / beta]])
            Rot = np.array([[np.cos(theta), -np.sin(theta)],
                            [np.sin(theta), np.cos(theta)]])
            position_w = np.dot(np.linalg.inv(Rot),
                                position_c - np.array([[tx], [ty]]))
            xw = position_w[0][0]
            yw = position_w[1][0]

            print(xw)
            print(yw)

            ################################# Your Code End Here #################################

            xy_w = str(xw) + str(' ') + str(yw)
            # print(xy_w)
            self.coord_pub.publish(xy_w)
Example #5
0
    def image_callback(self, data):

		global theta
		global beta
		global tx
		global ty

		try:
		    # Convert ROS image to OpenCV image
		    raw_image = self.bridge.imgmsg_to_cv2(data, "bgr8")
		except CvBridgeError as e:
		    print(e)

		# Flip the image 180 degrees
		cv_image = cv2.flip(raw_image, -1)

		# Draw a black line on the image
		cv2.line(cv_image, (0,50), (640,50), (0,0,0), 5)

		# cv_image is normal color image
		blob_image_center = blob_search(cv_image, self.detector)

		if(len(blob_image_center) == 0):
			print("No blob found!")
			self.coord_pub.publish("")
		else:
			x = int(blob_image_center[0].split()[0])
			y = int(blob_image_center[0].split()[1])
			# print("Blob found! ({0}, {1})".format(x, y))

			################################ Your Code Start Here ################################

			# Given theta, beta, tx, ty, calculate the world coordinate of x,y namely xw, yw  







			################################# Your Code End Here ################################# 

			xy_w = str(xw) + str(' ') + str(yw)
			# print(xy_w)
			self.coord_pub.publish(xy_w)
    def image_callback(self, data):
    	global theta
    	global beta
    	global tx
    	global ty

    	try:
    		raw_image = self.bridge.imgmsg_to_cv2(data, "bgr8")
    	except CvBridgeError as e:
    		print(e)

    	cv_image = cv2.flip(raw_image, -1)
    	cv2.line(cv_image, (0,50), (640,50), (0,0,0), 5)

    	blob_image_center = blob_search(cv_image, self.detector)

    	if(len(blob_image_center) == 0):
    		print("No blob found!")
    		self.coord_pub.publish("")
    	elif(len(blob_image_center) == 1):
    		x1 = int(blob_image_center[0][0])
    		y1 = int(blob_image_center[0][1])
    		# x2 = int(blob_image_center[1][0])
    		# y2 = int(blob_image_center[1][1])

    		# if x1 > x2:
    		# 	T = [x1,x2, y1, y2]
    		# 	x2 = T[0]
    		# 	x1 = T[1]
    		# 	y2 = T[2]
    		# 	y1 = T[3]

    		A = np.array([[x1/beta], [y1/beta]])
    		Rz = np.array([[math.cos(theta), -1*math.sin(theta)], [math.sin(theta), math.cos(theta)]])
    		B = Rz.dot(A) + np.array([[tx], [ty]])

    		xw = B[1,0]
    		yw = B[0,0]

    		xy_w = str(xw) + str(' ') + str(yw)
    		print(xy_w)
    		self.coord_pub.publish(xy_w)
Example #7
0
    def image_callback(self, data):

        global theta
        global beta
        global tx
        global ty

        try:
            # Convert ROS image to OpenCV image
            raw_image = self.bridge.imgmsg_to_cv2(data, "bgr8")
        except CvBridgeError as e:
            print(e)

        # Flip the image 180 degrees
        cv_image = cv2.flip(raw_image, -1)

        # Draw a black line on the image
        cv2.line(cv_image, (0, 50), (640, 50), (0, 0, 0), 5)

        # cv_image is normal color image
        blob_image_center, green_center, pink_center = blob_search(
            cv_image, self.detector)

        print(blob_image_center)

        if (len(blob_image_center) == 0):
            #print("No blob found!")
            self.coord_pub.publish("")
        else:

            x = int(blob_image_center[0].split()[1])
            y = int(blob_image_center[0].split()[0])
            xw = (x - 240) / beta - tx
            yw = (y - 320) / beta - ty

            xy_w = str(xw) + str(' ') + str(yw)
            print(xy_w)
            self.coord_pub.publish(xy_w)
    def image_callback(self, data):

        global theta
        global beta
        global tx
        global ty

        try:
            # Convert ROS image to OpenCV image
            raw_image = self.bridge.imgmsg_to_cv2(data, "bgr8")
        except CvBridgeError as e:
            print(e)

        # Flip the image 180 degrees
        cv_image = cv2.flip(raw_image, -1)

        # Draw a black line on the image
        cv2.line(cv_image, (0, 50), (640, 50), (0, 0, 0), 5)

        # cv_image is normal color image
        blob_image_center = blob_search(cv_image, self.detector)

        # Given world coordinate (xw, yw)
        xw = 0.2875
        yw = 0.1125

        # Only two blob center are found on the image
        if (len(blob_image_center) == 2):

            x1 = int(blob_image_center[0][0])
            y1 = int(blob_image_center[0][1])
            x2 = int(blob_image_center[1][0])
            y2 = int(blob_image_center[1][1])

            print("Blob Center 1: ({0}, {1}) and Blob Center 2: ({2}, {3})".
                  format(x1, y1, x2, y2))

            ################################# Your Code Start Here #################################

            # Calculate beta, tx and ty, given x1, y1, x2, y2

            d = ((x1 - x2)**2 + (y1 - y2)**2)**0.5

            if x1 > x2:
                T = [x1, x2, y1, y2]
                x2 = T[0]
                x1 = T[1]
                y2 = T[2]
                y1 = T[3]

            beta = d / 0.1  # Pixels per meter

            theta = math.asin((y2 - y1) / d)

            Rz = np.array([[math.cos(theta), -1 * math.sin(theta)],
                           [math.sin(theta), math.cos(theta)]])

            # Calculate Tx, Ty

            A = Rz.dot(np.array([[x1 / beta], [y1 / beta]]))

            cam_origin_x = yw - A[0, 0]
            cam_origin_y = xw - A[1, 0]

            tx = cam_origin_x
            ty = cam_origin_y

            ################################## Your Code End Here ##################################

            print("theta = {0}\nbeta = {1}\ntx = {2}\nty = {3}\n".format(
                theta, beta, tx, ty))

        else:
            print("No Blob found! ")
    def image_callback(self, data):

        global theta
        global beta
        global tx
        global ty

        try:
            # Convert ROS image to OpenCV image
            raw_image = self.bridge.imgmsg_to_cv2(data, "bgr8")
        except CvBridgeError as e:
            print(e)

        # Flip the image 180 degrees
        cv_image = cv2.flip(raw_image, -1)

        # Draw a black line on the image
        cv2.line(cv_image, (0, 50), (640, 50), (0, 0, 0), 5)

        # cv_image is normal color image
        blob_image_center = blob_search(cv_image, self.detector)

        # Given world coordinate (xw, yw)
        xw = 0.2875
        yw = 0.1125

        print(blob_image_center, " BLOB_TEST")

        # Only two blob center are found on the image
        if (len(blob_image_center) < 2):
            print("Not enough blobs found. " + str(len(blob_image_center)) +
                  " found.")

        elif (len(blob_image_center) == 2):

            c1 = int(blob_image_center[0].split()[0])
            r1 = int(blob_image_center[0].split()[1])
            c2 = int(blob_image_center[1].split()[0])
            r2 = int(blob_image_center[1].split()[1])

            print("Blob Center 1: ({0}, {1}) and Blob Center 2: ({2}, {3})".
                  format(c1, r1, c2, r2))

            ################################# Your Code Start Here #################################

            # Calculate beta, tx and ty, given c1, r1, c2, r2
            crop_top_row = 130
            crop_bottom_row = 350
            crop_top_col = 130
            crop_bottom_col = 450

            O_r = -(.5 * (crop_top_col - crop_bottom_col))  # O_r = .5*height
            O_c = -(.5 * (crop_top_row - crop_bottom_row))  # O_c = .5*width

            beta = 770
            theta = .019682  #radians

            tx = 7 / beta  #meters
            ty = 15 / beta  #meters
            X_c_1 = (r1 - O_r) / beta
            Y_c_1 = (c1 - O_c) / beta
            X_c_2 = (r2 - O_r) / beta
            Y_c_2 = (c2 - O_c) / beta

            rotation = np.array(
                [[cos(theta * np.pi / 180), -sin(theta * np.pi / 180)],
                 [sin(theta * np.pi / 180),
                  cos(theta * np.pi / 180)]])
            translation = np.array([[tx], [ty]])
            camera_1 = np.array([[X_c_1], [Y_c_1]])
            camera_2 = np.array([[X_c_2], [Y_c_2]])

            world_1 = np.dot(np.linalg.inv(translation),
                             np.subtract(camera_1 - translation))
            world_2 = np.dot(np.linalg.inv(translation),
                             np.subtract(camera_2 - translation))

            ################################## Your Code End Here ##################################

            print("theta = {0}\nbeta = {1}\ntx = {2}\nty = {3}\n".format(
                theta, beta, tx, ty))
        else:
            print("To many points returned " + str(len(blob_image_center)) +
                  " found.")
Example #10
0
    def image_callback(self, data):

        global theta
        global beta
        global tx
        global ty

        try:
            # Convert ROS image to OpenCV image
            raw_image = self.bridge.imgmsg_to_cv2(data, "bgr8")
        except CvBridgeError as e:
            print(e)

        # Flip the image 180 degrees
        cv_image = cv2.flip(raw_image, -1)

        # Draw a black line on the image
        cv2.line(cv_image, (0, 50), (640, 50), (0, 0, 0), 5)

        # cv_image is normal color image
        blob_image_center = blob_search(cv_image, self.detector)

        # Given world coordinate (xw, yw)
        xw = 0.2875
        yw = 0.1125

        # Only two blob center are found on the image
        if (len(blob_image_center) < 2):
            print("Not enough blobs found. " + str(len(blob_image_center)) +
                  " found.")

        elif (len(blob_image_center) == 2):

            c1 = int(blob_image_center[0].split()[0])
            r1 = int(blob_image_center[0].split()[1])
            c2 = int(blob_image_center[1].split()[0])
            r2 = int(blob_image_center[1].split()[1])

            print("Blob Center 1: ({0}, {1}) and Blob Center 2: ({2}, {3})".
                  format(c1, r1, c2, r2))

            ################################# Your Code Start Here #################################

            if c1 < c2:
                c = c1
                r = r1
            else:
                c = c2
                r = r2
            # Calculate beta, tx and ty, given c1, r1, c2, r2
            theta = np.arctan((r1 - r2) / (c1 - c2))
            theta = theta * 180 / np.pi
            beta = np.sqrt((c1 - c2)**2 + (r1 - r2)**2) / 0.1
            Ttmp = np.array([[0], [0]])
            xc_mat = np.array([[(r - 249) / beta * 1.0],
                               [(c - 369) / beta * 1.0]])
            xw_mat = np.array([[xw], [yw]])
            Rot = np.array([[np.cos(theta), -np.sin(theta)],
                            [np.sin(theta), np.cos(theta)]])
            Ttmp = xc_mat - np.dot(Rot, xw_mat)
            tx = Ttmp[0][0]
            ty = Ttmp[1][0]

            ################################## Your Code End Here ##################################

            print("theta = {0}\nbeta = {1}\ntx = {2}\nty = {3}\n".format(
                theta, beta, tx, ty))

        else:
            print("To many points returned " + str(len(blob_image_center)) +
                  " found.")
Example #11
0
    def image_callback(self, data):

        global theta
        global beta
        global tx
        global ty

        try:
            # Convert ROS image to OpenCV image
            raw_image = self.bridge.imgmsg_to_cv2(data, "bgr8")
        except CvBridgeError as e:
            print(e)

        # Flip the image 180 degrees
        cv_image = cv2.flip(raw_image, -1)

        # Draw a black line on the image
        cv2.line(cv_image, (0, 50), (640, 50), (0, 0, 0), 5)

        # cv_image is normal color image
        blob_image_center = blob_search(cv_image, self.detector)

        # Given world coordinate (xw, yw)
        xw = 0.2875
        yw = 0.1125

        # Only two blob center are found on the image
        if (len(blob_image_center) < 2):
            print("Not enough blobs found. and we are here " +
                  str(len(blob_image_center)) + " found.")

        elif (len(blob_image_center) == 2):

            c1 = int(blob_image_center[0].split()[0])
            r1 = int(blob_image_center[0].split()[1])
            c2 = int(blob_image_center[1].split()[0])
            r2 = int(blob_image_center[1].split()[1])

            print("Blob Center 1: ({0}, {1}) and Blob Center 2: ({2}, {3})".
                  format(c1, r1, c2, r2))

            ################################# Your Code Start Here #################################

            # Calculate beta, tx and ty, given c1, r1, c2, r2

            O_r = cv_image.shape[0] / 2
            O_c = cv_image.shape[1] / 2
            # O_c -= 200

            print("O_r = ", O_r, ", O_c = ", O_c)

            dist = ((c1 - c2)**2 + (r1 - r2)**2)**0.5

            beta = dist / 0.1

            theta = np.arcsin((r1 - r2) / dist)

            if (r1 < r2):
                r = r1
                c = c1
            else:
                r = r2
                c = c2

            x_c = (r - O_r) / beta
            y_c = (c - O_c) / beta

            tx = x_c - xw * np.cos(theta) + yw * np.sin(theta)
            ty = y_c - xw * np.sin(theta) - yw * np.cos(theta)

            ################################## Your Code End Here ##################################

            print("theta = {0}\nbeta = {1}\ntx = {2}\nty = {3}\n".format(
                theta, beta, tx, ty))

        else:
            print("To many points returned " + str(len(blob_image_center)) +
                  " found.")
Example #12
0
    def image_callback(self, data):

        global theta
        global beta
        global tx
        global ty

        try:
            # Convert ROS image to OpenCV image
            raw_image = self.bridge.imgmsg_to_cv2(data, "bgr8")
        except CvBridgeError as e:
            print(e)

        # Flip the image 180 degrees
        cv_image = cv2.flip(raw_image, -1)

        # Draw a black line on the image
        cv2.line(cv_image, (0,50), (640,50), (0,0,0), 5)

        # cv_image is normal color image
        blob_image_center = blob_search(cv_image, self.detector)

        if(len(blob_image_center) == 0):
            print("No blob found!")
            self.coord_pub.publish("")
        else:
            c = int(blob_image_center[0].split()[0]) # - 98
            r = int(blob_image_center[0].split()[1])
            # print("Blob found! ({0}, {1})".format(c, r))

            ################################ Your Code Start Here ################################

            # Given theta, beta, tx, ty, calculate the world coordinate of c,r namely xw, yw  

            O_r = cv_image.shape[0] / 2
            O_c = cv_image.shape[1] / 2

            print("O_r = ", O_r, ", O_c = ", O_c)
            
            x_c = (r - O_r) / beta
            y_c = (c - O_c) / beta

            R = np.array([[np.cos(theta), -np.sin(theta)], [np.sin(theta), np.cos(theta)]])
            R_inv = np.linalg.inv(R)

            P_c = np.array([[x_c], [y_c]])
            T = np.array([[tx], [ty]])

            P_w = np.matmul(R_inv, (P_c - T))

            xw = P_w[0][0]
            yw = P_w[1][0]

            # xw = x_c - tx
            # yw = y_c - ty

            ################################# Your Code End Here ################################# 

            xy_w = str(xw) + str(' ') + str(yw)
            print(xy_w)
            self.coord_pub.publish(xy_w)
Example #13
0
    def image_callback(self, data):

        global theta
        global beta
        global tx
        global ty

        try:
            # Convert ROS image to OpenCV image
            raw_image = self.bridge.imgmsg_to_cv2(data, "bgr8")
        except CvBridgeError as e:
            print(e)

        # Flip the image 180 degrees
        cv_image = cv2.flip(raw_image, -1)

        # Draw a black line on the image
        cv2.line(cv_image, (0,50), (640,50), (0,0,0), 5)

        # cv_image is normal color image
        blob_image_center = blob_search(cv_image, self.detector)

        if(len(blob_image_center) == 0):
            print("No blob found!")
            self.coord_pub.publish("")
        else:
            c = float(blob_image_center[0].split()[0])
            r = float(blob_image_center[0].split()[1])
            print("Blob found! ({0}, {1})".format(str(c), str(r)))

            ################################ Your Code Start Here ################################

            # Given theta, beta, tx, ty, calculate the world coordinate of c,r namely xw, yw
            beta = 770
            theta = .019682 #radians

            tx = -0.272974314536
            ty = -0.200657008695

            X_c_1 = (r - 220)/beta      # width = 220
            Y_c_1 = (c - 320)/beta      # height = 320

            rotation = np.array([[math.cos(theta), -math.sin(theta)],[math.sin(theta),math.cos(theta)]])
            translation = np.array([[tx],[ty]])
            camera_1 = np.array([[X_c_1],[Y_c_1]])

            world_1 = np.matmul(np.linalg.inv(rotation),(camera_1 - translation))
            xw = world_1[0][0]
            yw = world_1[1][0]

            # R = np.array([[np.cos(theta),-np.sin(theta)],[np.sin(theta),np.cos(theta)]])
            # Or = 240
            # Oc = 320
            # xc = (r-Or)/beta
            # yc = (c-Oc)/beta

            # C = np.array([[xc],[yc]])
            # T = np.array([[tx],[ty]])
            # R_inv = np.linalg.inv(R)
            # M = C-T
            # W = np.matmul(R_inv,M)
            # xw = W[0][0]
            # yw = W[1][0]

            #print(world_1, "WORLD_1")
            #print(type(world_1), "WORLD_1 FORMAT")

            ################################# Your Code End Here #################################

            xy_w = str(xw) + str(' ') + str(yw)
            print(xy_w)
            self.coord_pub.publish(xy_w)