Bump samples to Zivid SDK 2.13

This commit adds the following changes: New samples: - network_configuration - automatic_network_configuration_for_cameras - transform_point_cloud_via_aruco_marker - transform_point_cloud_via_checkerboard - roi_box_via_aruco_marker Modification of samples: - hand-eye sample now supports ArUco markers - in various samples, calibration.detect_calibration_board() is used instead of experimental.calibration.detect_feature_points(). This removes the zivid.experimental.calibration dependency.
zivid · Jul 30, 2024 · 2111652 · 2111652
1 parent 19cae6b
commit 2111652
Show file tree

Hide file tree

Showing 24 changed files with 1,095 additions and 237 deletions.
diff --git a/README.md b/README.md
@@ -1,6 +1,6 @@
 # Python samples
 
-This repository contains python code samples for Zivid SDK v2.12.0. For
+This repository contains python code samples for Zivid SDK v2.13.1. For
 tested compatibility with earlier SDK versions, please check out
 [accompanying
 releases](https://github.com/zivid/zivid-python-samples/tree/master/../../releases).
@@ -60,6 +60,9 @@ from the camera can be used.
           - [capture\_hdr\_print\_normals](https://github.com/zivid/zivid-python-samples/tree/master/source/camera/advanced/capture_hdr_print_normals.py) - Capture Zivid point clouds, compute normals and print a
             subset.
       - **info\_util\_other**
+          - [automatic\_network\_configuration\_for\_cameras](https://github.com/zivid/zivid-python-samples/tree/master/source/camera/info_util_other/automatic_network_configuration_for_cameras.py) - Automatically set the IP addresses of any number of
+            cameras to be in the same subnet as the provided IP address
+            of the network interface.
           - [camera\_info](https://github.com/zivid/zivid-python-samples/tree/master/source/camera/info_util_other/camera_info.py) - Print version information for Python, zivid-python and
             Zivid SDK, then list cameras and print camera info and state
             for each connected camera.
@@ -69,6 +72,8 @@ from the camera can be used.
           - [firmware\_updater](https://github.com/zivid/zivid-python-samples/tree/master/source/camera/info_util_other/firmware_updater.py) - Update firmware on the Zivid camera.
           - [get\_camera\_intrinsics](https://github.com/zivid/zivid-python-samples/tree/master/source/camera/info_util_other/get_camera_intrinsics.py) - Read intrinsic parameters from the Zivid camera (OpenCV
             model) or estimate them from the point cloud.
+          - [network\_configuration](https://github.com/zivid/zivid-python-samples/tree/master/source/camera/info_util_other/network_configuration.py) - Uses Zivid API to change the IP address of the Zivid
+            camera.
           - [warmup](https://github.com/zivid/zivid-python-samples/tree/master/source/camera/info_util_other/warmup.py) - A basic warm-up method for a Zivid camera with specified
             time and capture cycle.
       - **maintenance**
@@ -114,8 +119,16 @@ from the camera can be used.
             images to find the marker coordinates (2D and 3D).
           - [reproject\_points](https://github.com/zivid/zivid-python-samples/tree/master/source/applications/advanced/reproject_points.py) - Illuminate checkerboard (Zivid Calibration Board) corners
             by getting checkerboard pose
+          - [roi\_box\_via\_aruco\_marker](https://github.com/zivid/zivid-python-samples/tree/master/source/applications/advanced/roi_box_via_aruco_marker.py) - Filter the point cloud based on a ROI box given relative
+            to the ArUco marker on a Zivid Calibration Board.
           - [roi\_box\_via\_checkerboard](https://github.com/zivid/zivid-python-samples/tree/master/source/applications/advanced/roi_box_via_checkerboard.py) - Filter the point cloud based on a ROI box given relative
             to the Zivid Calibration Board.
+          - [transform\_point\_cloud\_via\_aruco\_marker](https://github.com/zivid/zivid-python-samples/tree/master/source/applications/advanced/transform_point_cloud_via_aruco_marker.py) - Transform a point cloud from camera to ArUco marker
+            coordinate frame by estimating the marker's pose from the
+            point cloud.
+          - [transform\_point\_cloud\_via\_checkerboard](https://github.com/zivid/zivid-python-samples/tree/master/source/applications/advanced/transform_point_cloud_via_checkerboard.py) - Transform a point cloud from camera to checkerboard (Zivid
+            Calibration Board) coordinate frame by getting checkerboard
+            pose from the API.
           - **hand\_eye\_calibration**
               - [pose\_conversions](https://github.com/zivid/zivid-python-samples/tree/master/source/applications/advanced/hand_eye_calibration/pose_conversions.py) - Convert to/from Transformation Matrix (Rotation Matrix
                 + Translation Vector).
@@ -134,7 +147,8 @@ from the camera can be used.
       - [display](https://github.com/zivid/zivid-python-samples/tree/master/source/sample_utils/display.py) - Display relevant data for Zivid Samples.
       - [paths](https://github.com/zivid/zivid-python-samples/tree/master/source/sample_utils/paths.py) - Get relevant paths for Zivid Samples.
       - [robodk\_tools](https://github.com/zivid/zivid-python-samples/tree/master/source/sample_utils/robodk_tools.py) - Robot Control Module
-      - [save\_load\_matrix](https://github.com/zivid/zivid-python-samples/tree/master/source/sample_utils/save_load_matrix.py) - try:
+      - [save\_load\_matrix](https://github.com/zivid/zivid-python-samples/tree/master/source/sample_utils/save_load_matrix.py) - Save and load Zivid 4x4 transformation matrices from and to
+        YAML files.
       - [white\_balance\_calibration](https://github.com/zivid/zivid-python-samples/tree/master/source/sample_utils/white_balance_calibration.py) - Balance color for 2D capture using white surface as reference.
   - **applications**
       - **advanced**

diff --git a/continuous-integration/setup.sh b/continuous-integration/setup.sh
@@ -28,7 +28,7 @@ function install_www_deb {
     rm -r $TMP_DIR || exit
 }
 
-install_www_deb "https://downloads.zivid.com/sdk/releases/2.12.0+6afd4961-1/u${VERSION_ID:0:2}/zivid_2.12.0+6afd4961-1_amd64.deb" || exit
+install_www_deb "https://downloads.zivid.com/sdk/releases/2.13.1+18e79e79-1/u${VERSION_ID:0:2}/zivid_2.13.1+18e79e79-1_amd64.deb" || exit
 
 python3 -m pip install --upgrade pip || exit
 python3 -m pip install --requirement "$ROOT_DIR/requirements.txt" || exit

diff --git a/source/applications/advanced/auto_2d_settings.py b/source/applications/advanced/auto_2d_settings.py
@@ -16,8 +16,6 @@
 first. If you want to use your own white reference (white wall, piece of paper, etc.) instead of using the calibration
 board, you can provide your own mask in _main(). Then you will have to specify the lower limit for f-number yourself.
 
-Note: This example uses experimental SDK features, which may be modified, moved, or deleted in the future without notice.
-
 """
 
 import argparse
@@ -29,7 +27,6 @@
 import matplotlib.pyplot as plt
 import numpy as np
 import zivid
-import zivid.experimental.calibration
 from sample_utils.calibration_board_utils import find_white_mask_from_checkerboard
 from sample_utils.white_balance_calibration import compute_mean_rgb_from_mask, white_balance_calibration
 
@@ -152,7 +149,7 @@ def _find_white_mask_and_distance_to_checkerboard(camera: zivid.Camera) -> Tuple
         settings = _capture_assistant_settings(camera)
         frame = camera.capture(settings)
 
-        checkerboard_pose = zivid.experimental.calibration.detect_feature_points(frame).pose().to_matrix()
+        checkerboard_pose = zivid.calibration.detect_calibration_board(frame).pose().to_matrix()
         distance_to_checkerboard = checkerboard_pose[2, 3]
 
         rgb = frame.point_cloud().copy_data("rgba")[:, :, :3]
@@ -520,10 +517,10 @@ def _print_poor_pixel_distribution(rgb: np.ndarray) -> None:
     black_and = np.sum(np.logical_and(np.logical_and(rgb[:, :, 0] == 0, rgb[:, :, 1] == 0), rgb[:, :, 2] == 0))
 
     print("Distribution of saturated (255) and black (0) pixels with final settings:")
-    print(f"Saturated pixels (at least one channel): {saturated_or}\t ({100*saturated_or/total_num_pixels:.2f}%)")
-    print(f"Saturated pixels (all channels):\t {saturated_and}\t ({100*saturated_and/total_num_pixels:.2f}%)")
-    print(f"Black pixels (at least one channel):\t {black_or}\t ({100*black_or/total_num_pixels:.2f}%)")
-    print(f"Black pixels (all channels):\t\t {black_and}\t ({100*black_and/total_num_pixels:.2f}%)")
+    print(f"Saturated pixels (at least one channel): {saturated_or}\t ({100 * saturated_or / total_num_pixels:.2f}%)")
+    print(f"Saturated pixels (all channels):\t {saturated_and}\t ({100 * saturated_and / total_num_pixels:.2f}%)")
+    print(f"Black pixels (at least one channel):\t {black_or}\t ({100 * black_or / total_num_pixels:.2f}%)")
+    print(f"Black pixels (all channels):\t\t {black_and}\t ({100 * black_and / total_num_pixels:.2f}%)")
 
 
 def _plot_image_with_histogram(rgb: np.ndarray, settings_2d: zivid.Settings2D) -> None:

diff --git a/source/applications/advanced/get_checkerboard_pose_from_zdf.py b/source/applications/advanced/get_checkerboard_pose_from_zdf.py
@@ -5,16 +5,13 @@
 The checkerboard point cloud is also visualized with a coordinate system.
 The ZDF file for this sample can be found under the main instructions for Zivid samples.
 
-Note: This example uses experimental SDK features, which may be modified, moved, or deleted in the future without notice.
-
 """
 
 from pathlib import Path
 
 import numpy as np
 import open3d as o3d
 import zivid
-import zivid.experimental.calibration
 from sample_utils.paths import get_sample_data_path
 from sample_utils.save_load_matrix import assert_affine_matrix_and_save
 
@@ -35,8 +32,8 @@ def _create_open3d_point_cloud(point_cloud: zivid.PointCloud) -> o3d.geometry.Po
     xyz = np.nan_to_num(xyz).reshape(-1, 3)
     rgb = rgba[:, :, 0:3].reshape(-1, 3)
 
-    point_cloud_open3d = o3d.geometry.PointCloud(o3d.utility.Vector3dVector(xyz))
-    point_cloud_open3d.colors = o3d.utility.Vector3dVector(rgb / 255)
+    point_cloud_open3d = o3d.geometry.PointCloud(o3d.utility.Vector3dVector(xyz.astype(np.float64)))
+    point_cloud_open3d.colors = o3d.utility.Vector3dVector(rgb.astype(np.float64) / 255)
 
     refined_point_cloud_open3d = o3d.geometry.PointCloud.remove_non_finite_points(
         point_cloud_open3d, remove_nan=True, remove_infinite=True
@@ -75,9 +72,7 @@ def _main() -> None:
         point_cloud = frame.point_cloud()
 
         print("Detecting checkerboard and estimating its pose in camera frame")
-        transform_camera_to_checkerboard = (
-            zivid.experimental.calibration.detect_feature_points(frame).pose().to_matrix()
-        )
+        transform_camera_to_checkerboard = zivid.calibration.detect_calibration_board(frame).pose().to_matrix()
         print(f"Camera pose in checkerboard frame:\n{transform_camera_to_checkerboard}")
 
         transform_file_name = "CameraToCheckerboardTransform.yaml"

diff --git a/source/applications/advanced/hand_eye_calibration/hand_eye_calibration.py b/source/applications/advanced/hand_eye_calibration/hand_eye_calibration.py
@@ -1,17 +1,14 @@
 """
 Perform Hand-Eye calibration.
 
-Note: This example uses experimental SDK features, which may be modified, moved, or deleted in the future without notice.
-
 """
 
 import datetime
 from pathlib import Path
-from typing import List
+from typing import List, Tuple
 
 import numpy as np
 import zivid
-import zivid.experimental.calibration
 from sample_utils.save_load_matrix import assert_affine_matrix_and_save
 
 
@@ -26,7 +23,7 @@ def _enter_robot_pose(index: int) -> zivid.calibration.Pose:
 
     """
     inputted = input(
-        f"Enter pose with id={index} (a line with 16 space separated values describing 4x4 row-major matrix):"
+        f"Enter pose with id={index} (a line with 16 space separated values describing 4x4 row-major matrix): "
     )
     elements = inputted.split(maxsplit=15)
     data = np.array(elements, dtype=np.float64).reshape((4, 4))
@@ -46,14 +43,14 @@ def _perform_calibration(hand_eye_input: List[zivid.calibration.HandEyeInput]) -
 
     """
     while True:
-        calibration_type = input("Enter type of calibration, eth (for eye-to-hand) or eih (for eye-in-hand):").strip()
+        calibration_type = input("Enter type of calibration, eth (for eye-to-hand) or eih (for eye-in-hand): ").strip()
         if calibration_type.lower() == "eth":
-            print("Performing eye-to-hand calibration")
+            print(f"Performing eye-to-hand calibration with {len(hand_eye_input)} dataset pairs")
             print("The resulting transform is the camera pose in robot base frame")
             hand_eye_output = zivid.calibration.calibrate_eye_to_hand(hand_eye_input)
             return hand_eye_output
         if calibration_type.lower() == "eih":
-            print("Performing eye-in-hand calibration")
+            print(f"Performing eye-in-hand calibration with {len(hand_eye_input)} dataset pairs")
             print("The resulting transform is the camera pose in flange (end-effector) frame")
             hand_eye_output = zivid.calibration.calibrate_eye_in_hand(hand_eye_input)
             return hand_eye_output
@@ -78,6 +75,58 @@ def _assisted_capture(camera: zivid.Camera) -> zivid.Frame:
     return camera.capture(settings)
 
 
+def _handle_add_pose(
+    current_pose_id: int, hand_eye_input: List, camera: zivid.Camera, calibration_object: str
+) -> Tuple[int, List]:
+    """Acquire frame with capture assistant.
+
+    Args:
+        current_pose_id: Counter of the current pose in the hand-eye calibration dataset
+        hand_eye_input: List of hand-eye calibration dataset pairs (poses and point clouds)
+        camera: Zivid camera
+        calibration_object: m (for ArUco marker(s)) or c (for Zivid checkerboard)
+
+    Returns:
+        Tuple[int, List]: Updated current_pose_id and hand_eye_input
+
+    """
+
+    robot_pose = _enter_robot_pose(current_pose_id)
+
+    print("Detecting calibration object in point cloud")
+
+    if calibration_object == "c":
+
+        frame = zivid.calibration.capture_calibration_board(camera)
+        detection_result = zivid.calibration.detect_calibration_board(frame)
+
+        if detection_result.valid():
+            print("Calibration board detected")
+            hand_eye_input.append(zivid.calibration.HandEyeInput(robot_pose, detection_result))
+            current_pose_id += 1
+        else:
+            print("Failed to detect calibration board, ensure that the entire board is in the view of the camera")
+    elif calibration_object == "m":
+
+        frame = _assisted_capture(camera)
+
+        marker_dictionary = zivid.calibration.MarkerDictionary.aruco4x4_50
+        marker_ids = [1, 2, 3]
+
+        print(f"Detecting arUco marker IDs {marker_ids} from the dictionary {marker_dictionary}")
+        detection_result = zivid.calibration.detect_markers(frame, marker_ids, marker_dictionary)
+
+        if detection_result.valid():
+            print(f"ArUco marker(s) detected: {len(detection_result.detected_markers())}")
+            hand_eye_input.append(zivid.calibration.HandEyeInput(robot_pose, detection_result))
+            current_pose_id += 1
+        else:
+            print(
+                "Failed to detect any ArUco markers, ensure that at least one ArUco marker is in the view of the camera"
+            )
+    return current_pose_id, hand_eye_input
+
+
 def _main() -> None:
     app = zivid.Application()
 
@@ -88,31 +137,26 @@ def _main() -> None:
     hand_eye_input = []
     calibrate = False
 
+    while True:
+        calibration_object = input(
+            "Enter calibration object you are using, m (for ArUco marker(s)) or c (for Zivid checkerboard): "
+        ).strip()
+        if calibration_object.lower() == "m" or calibration_object.lower() == "c":
+            break
+
     print(
         "Zivid primarily operates with a (4x4) transformation matrix. To convert\n"
         "from axis-angle, rotation vector, roll-pitch-yaw, or quaternion, check out\n"
         "our pose_conversions sample."
     )
 
     while not calibrate:
-        command = input("Enter command, p (to add robot pose) or c (to perform calibration):").strip()
+        command = input("Enter command, p (to add robot pose) or c (to perform calibration): ").strip()
         if command == "p":
             try:
-                robot_pose = _enter_robot_pose(current_pose_id)
-
-                frame = _assisted_capture(camera)
-
-                print("Detecting checkerboard in point cloud")
-                detection_result = zivid.experimental.calibration.detect_feature_points(frame)
-
-                if detection_result.valid():
-                    print("Calibration board detected")
-                    hand_eye_input.append(zivid.calibration.HandEyeInput(robot_pose, detection_result))
-                    current_pose_id += 1
-                else:
-                    print(
-                        "Failed to detect calibration board, ensure that the entire board is in the view of the camera"
-                    )
+                current_pose_id, hand_eye_input = _handle_add_pose(
+                    current_pose_id, hand_eye_input, camera, calibration_object
+                )
             except ValueError as ex:
                 print(ex)
         elif command == "c":