The Obi Zhongguang Astra depth camera directly displays RGB color images, depth images, and IR infrared images through OpenNI.

The premise of this article is that the astra depth camera driver has been installed and can be used normally. If it is not installed properly, please read another article.

Use the Astra depth camera on the Thinker Board 2 development board.
You can also see other people’s tutorials similar to the Astra depth camera.

Create a python file and copy the following code to use it.
Note that the depth map and ir map conflict and cannot be used at the same time.

from openni import openni2
import numpy as np
import cv2


if __name__ == "__main__":

    openni2.initialize()
    #设备型号
    dev = openni2.Device.open_any()
    #打印设备型号
    print(dev.get_device_info())

    #创建深度流的通道
    depth_stream = dev.create_depth_stream()
    depth_stream.start()
    #彩色通道
    color_stream = dev.create_color_stream()
    color_stream.start()
    #红外通道
    #ir_stream = dev.create_ir_stream()
    #ir_stream.start()



    #深度窗口
    cv2.namedWindow('depth')
    #彩色窗口
    cv2.namedWindow('color')
    #红外窗口
    #cv2.namedWindow('ir')

    capture_flag = 0

    base_path = '../'

    count = 1
    #设置视频的分辨率和长宽以及fps等等内容
    # ir_stream.set_video_mode(c_api.OniVideoMode(pixelFormat = c_api.OniPixelFormat.ONI_PIXEL_FORMAT_GRAY16, resolutionX = 320, resolutionY = 240, fps = 30))

    while True:
    # 显示深度图像
        dframe = depth_stream.read_frame()
        # 转换数据格式
        dframe_data = np.array(dframe.get_buffer_as_triplet()).reshape([480, 640, 2])
        dpt1 = np.asarray(dframe_data[:, :, 0], dtype='float32')
        dpt2 = np.asarray(dframe_data[:, :, 1], dtype='float32')

        dpt2 *= 255
        # 对于为什么要乘于255的解答
        # 深度图像的深度值 是按照16位长度（两字节）的数据格式存储的，也可以认为前八位是高字节，后八位是低字节。
        # 因此一张深度图像如果是 640480分辨率的话，那么图像字节大小 就是 640480*2，其中一个字节是8位（255）
        dpt = dpt1 + dpt2
        # 转换一下，重要，陈20221222
        dpt = dpt.astype(np.uint16)

        # cv2里面的函数，就是类似于一种筛选
        '假设我们需要让我们的深度摄像头感兴趣的距离范围有差别地显示，那么我们就需要确定一个合适的alpha值，公式为：有效距离*alpha=255，' \
        '假设我们想让深度摄像头8m距离内的深度被显示，>8m的与8m的颜色显示相同，那么alpha=255/(8*10^3)≈0.03，' \
        '假设我们想让深度摄像头6m距离内的深度被显示，>6m的与6m的颜色显示相同，那么alpha=255/(6*10^3)≈0.0425'
        dim_gray = cv2.convertScaleAbs(dpt, alpha=0.17)
        # 对深度图像进行一种图像的渲染，目前有11种渲染方式，大家可以逐一去试下
        depth_colormap = cv2.applyColorMap(dim_gray, 2)  # 有0~11种渲染的模式
        cv2.imshow('depth', depth_colormap)

    # 显示RGB图像
        cframe = color_stream.read_frame()
        cframe_data = np.array(cframe.get_buffer_as_triplet()).reshape([480, 640, 3])
        R = cframe_data[:, :, 0]
        G = cframe_data[:, :, 1]
        B = cframe_data[:, :, 2]
        cframe_data = np.transpose(np.array([B, G, R]), [1, 2, 0])
        # print(cframe_data.shape)
        cv2.imshow('color', cframe_data)

    # 显示ir图像
    #     iframe = ir_stream.read_frame()
    #     iframe_data = iframe.get_buffer_as_uint16()
    #     img = np.frombuffer(iframe_data, dtype=np.uint16)
    #     img.shape = (480, 640)
    #     img = img.astype(np.float) / 1024
    #     cv2.imshow("ir", img)

        key = cv2.waitKey(30)
        if int(key) == ord('q'):
            break

        if int(key) == ord('r'):
            capture_flag = 1
        if capture_flag == 1:
            name = str(count).zfill(8)
            cv2.imwrite(base_path + 'img/' + name + ".jpg", depth_colormap)
            #cv2.imwrite(base_path + 'img/' + name + ".jpg", cframe_data)
            # cv2.imwrite(base_path + 'img/' + name + ".jpg", img)
            count = count + 1

    depth_stream.stop()
    color_stream.stop()
    #ir_stream.stop()
    dev.close()

I won’t show the ir diagram here.