Skip to content

Latest commit

 

History

History
505 lines (350 loc) · 14.7 KB

README.md

File metadata and controls

505 lines (350 loc) · 14.7 KB
Raw

Camera

Introduction

The camera_standard repository enables your application to provide camera functions. You can access and operate the camera device and develop new functions through open APIs. Common operations include preview, photographing and video recording.

Basic Concepts

  • Capture Photo

    This function is used to capture a photo.

  • Preview

    This function repeatedly captures multiple frames in the buffer once the camera is started. It is used to preview the subject before capturing a photo or start recording a video.

  • Record Video

    This function repeatedly captures multiple frames in the buffer once the recording is started and continues till the recording is stopped. It is used for video recording.

Figure 1 Camera architecture

Directory Structure

The structure of the repository directory is as follows:

/foundation/multimedia/camera_framework  # Camera code
├── frameworks                          # Framework code
│   ├── native                          # Internal Native API Implementation
│   │   ├── camera                      # Camera Framework Implementation
│   │   └── metadata                    # Metadata Implementation
│   └── js                              # External JS API Implementation
│       └── camera_napi                 # Camera NAPI Implementation
├── interfaces                          # Interfaces
│   ├── inner_api                       # Internal Native APIs
│   └── kits                            # External JS APIs
├── LICENSE                             # License file
├── ohos.build                          # Build file
├── sa_profile                          # Service configuration profile
└── services                            # Service code
    ├── camera_service                  # Camera Service Implementation
    └── etc                             # Camera Service Init Configuration

Usage Guidelines

To Capture a Photo

Steps to capture a photo:

  1. Create the buffer consumer listener(CaptureSurfaceListener) to save the image.

    class CaptureSurfaceListener : public IBufferConsumerListener {
public:
    int32_t mode_;
    sptr<Surface> surface_;

    void OnBufferAvailable() override
    {
        int32_t flushFence = 0;
        int64_t timestamp = 0;
        OHOS::Rect damage; // initialize the damage

        OHOS::sptr<OHOS::SurfaceBuffer> buffer = nullptr;
        surface_->AcquireBuffer(buffer, flushFence, timestamp, damage);
        if (buffer != nullptr) {
            void* addr = buffer->GetVirAddr();
            int32_t size = buffer->GetSize();

            // Save the buffer(addr) to a file.

            surface_->ReleaseBuffer(buffer, -1);
        }
    }
};

  2. Get the Camera Manager Instance and retrieve the camera object list.

    sptr<CameraManager> camManagerObj = CameraManager::GetInstance();
std::vector<sptr<CameraInfo>> cameraObjList = camManagerObj->GetCameras();

  3. Create the Camera Input using the Camera Object to open the camera.

    sptr<CaptureInput> cameraInput = camManagerObj->CreateCameraInput(cameraObjList[0]);

  4. Create the Capture Session

    sptr<CaptureSession> captureSession = camManagerObj->CreateCaptureSession();

  5. Begin configuring the Capture Session.

    int32_t result = captureSession->BeginConfig();

  6. Add the Camera Input to the Capture Session.

    result = captureSession->AddInput(cameraInput);

  7. Create the Consumer Surface and register listerner for buffer updates. The photo width and height can be configured to the supported resolution which is 1280x960.

    sptr<Surface> photoSurface = Surface::CreateSurfaceAsConsumer();
int32_t photoWidth = 1280;
int32_t photoHeight = 960;
photoSurface->SetDefaultWidthAndHeight(photoWidth, photoHeight);
photoSurface->SetUserData(CameraManager::surfaceFormat, std::to_string(OHOS_CAMERA_FORMAT_JPEG));
sptr<CaptureSurfaceListener> capturelistener = new(std::nothrow) CaptureSurfaceListener();
capturelistener->mode_ = MODE_PHOTO;
capturelistener->surface_ = photoSurface;
photoSurface->RegisterConsumerListener((sptr<IBufferConsumerListener> &)capturelistener);

  8. Create a Photo Output with Surface created above.

    sptr<CaptureOutput> photoOutput = camManagerObj->CreatePhotoOutput(photoSurface);

  9. Add the Photo Output to the Capture Session.

    result = captureSession->AddOutput(photoOutput);

  10. Commit the configuration to Capture Session.

    result = captureSession->CommitConfig();

  11. Capture the Photo.

    result = ((sptr<PhotoOutput> &)photoOutput)->Capture();

  12. Release the Capture Session resources.

    captureSession->Release();

  13. Release the Camera Input to close the camera.

    cameraInput->Release();

To Start and Stop Preview

Steps to start and stop preview:

  1. Get the Camera Manager Instance and retrieve the camera object list.

    sptr<CameraManager> camManagerObj = CameraManager::GetInstance();
std::vector<sptr<CameraInfo>> cameraObjList = camManagerObj->GetCameras();

  2. Create the Camera Input using the Camera Object to open the camera.

    sptr<CaptureInput> cameraInput = camManagerObj->CreateCameraInput(cameraObjList[0]);

  3. Create the Capture Session

    sptr<CaptureSession> captureSession = camManagerObj->CreateCaptureSession();

  4. Begin configuring the Capture Session.

    int32_t result = captureSession->BeginConfig();

  5. Add the Camera Input to the Capture Session.

    result = captureSession->AddInput(cameraInput);

  6. Create a Preview Output with Surface obtained from Window Manager to render on display. The preview width and Height can be configured to one of the camera supported resolutions. And to save to a file, can follow the step to create Surface and register listerner for buffer updates mentioned in capture photo.

    int32_t previewWidth = 640;
int32_t previewHeight = 480;
previewSurface->SetUserData(CameraManager::surfaceFormat, std::to_string(OHOS_CAMERA_FORMAT_YCRCB_420_SP));
sptr<CaptureOutput> previewOutput = camManagerObj->CreateCustomPreviewOutput(previewSurface, previewWidth, previewHeight);

  7. Add the Preview Output to the Capture Session.

    result = captureSession->AddOutput(previewOutput);

  8. Commit the configuration to Capture Session.

    result = captureSession->CommitConfig();

  9. Start the Preview

    result = captureSession->Start();

  10. Stop the Preview when needed.

    result = captureSession->Stop();

  11. Release the Capture Session resources.

    captureSession->Release();

  12. Release the Camera Input to close the camera.

    cameraInput->Release();

To Record Video

Steps to record Video:

  1. Get the Camera Manager Instance and retrieve the camera object list.

    sptr<CameraManager> camManagerObj = CameraManager::GetInstance();
std::vector<sptr<CameraInfo>> cameraObjList = camManagerObj->GetCameras();

  2. Create the Camera Input using the Camera Object to open the camera.

    sptr<CaptureInput> cameraInput = camManagerObj->CreateCameraInput(cameraObjList[0]);

  3. Create the Capture Session

    sptr<CaptureSession> captureSession = camManagerObj->CreateCaptureSession();

  4. Begin configuring the Capture Session.

s int32_t result = captureSession->BeginConfig(); ```

  1. Add the Camera Input to the Capture Session.

    result = captureSession->AddInput(cameraInput);

  2. Create a Video Output with Surface obtained from Recoder to MUX with audio and save the file. And to save just Video buffer to a file, can follow the step to create Surface and register listerner for buffer updates mentioned in capture photo. The video resolution can be configured to one of the camera and recorder supported resolutions while setting the recorder configurations.

    videoSurface->SetUserData(CameraManager::surfaceFormat, std::to_string(OHOS_CAMERA_FORMAT_YCRCB_420_SP));
sptr<CaptureOutput> videoOutput = camManagerObj->CreateVideoOutput(videoSurface);

  3. Add the Video Output to the Capture Session.

    result = captureSession->AddOutput(videoOutput);

  4. Commit the configuration to Capture Session.

    result = captureSession->CommitConfig();

  5. Start the Video Recording.

    result = ((sptr<VideoOutput> &)videoOutput)->Start();

  6. Stop the recording when needed.

    result = ((sptr<VideoOutput> &)videoOutput)->Stop();

  7. Release the Capture Session resources.

    captureSession->Release();

  8. Release the Camera Input to close the camera.

    cameraInput->Release();

To Switch Between Camera Devices

Below steps demonstrate how to switch between the camera devices. Initially a video output is added to the capture session. If user wish to switch between cameras, existing input and output have to be removed first and then add a new input and output(i.e., photo output in this case).

  1. Get the Camera Manager Instance and retrieve the camera object list.

    sptr<CameraManager> camManagerObj = CameraManager::GetInstance();
std::vector<sptr<CameraInfo>> cameraObjList = camManagerObj->GetCameras();

  2. Create the Camera Input using the Camera Object to open the camera.

    sptr<CaptureInput> cameraInput = camManagerObj->CreateCameraInput(cameraObjList[0]);

  3. Create the Capture Session

    sptr<CaptureSession> captureSession = camManagerObj->CreateCaptureSession();

  4. Begin configuring the Capture Session.

    int32_t result = captureSession->BeginConfig();

  5. Add the Camera Input to the Capture Session.

    result = captureSession->AddInput(cameraInput);

  6. Create a Video Output with Surface

    sptr<CaptureOutput> videoOutput = camManagerObj->CreateVideoOutput(videoSurface);

  7. Add the Video Output to the Capture Session.

    result = captureSession->AddOutput(videoOutput);

  8. Commit the configuration to Capture Session.

    result = captureSession->CommitConfig();

  9. Start the Video Recording.

    result = ((sptr<VideoOutput> &)videoOutput)->Start();

  10. Stop the recording when needed.

    result = ((sptr<VideoOutput> &)videoOutput)->Stop();

  11. In order to remove camera input and output, configure the Capture Session again.

    int32_t result = captureSession->BeginConfig();

  12. Remove the Camera Input in the new capture session configuration.

    int32_t result = captureSession->RemoveInput(cameraInput);

  13. Remove the Camera Output as well.

    int32_t result = captureSession->RemoveOutut(videoOutput);

  14. Create new camera input, add it to capture session

    sptr<CaptureInput> cameraInput2 = camManagerObj->CreateCameraInput(cameraObjList[1]);
result = captureSession->AddInput(cameraInput2);

  15. Create the photo output and add the photo output to the capture session once it is successfully created. Create the Consumer Surface and register listerner for buffer updates prior to creation of photo output. The surface is used for creation of photo output.

    // Get the surface
sptr<Surface> photoSurface = Surface::CreateSurfaceAsConsumer();
int32_t photoWidth = 1280;
int32_t photoHeight = 960;
photoSurface->SetDefaultWidthAndHeight(photoWidth, photoHeight);
photoSurface->SetUserData(CameraManager::surfaceFormat, std::to_string(OHOS_CAMERA_FORMAT_JPEG));
sptr<CaptureSurfaceListener> capturelistener = new(std::nothrow) CaptureSurfaceListener();
capturelistener->mode_ = MODE_PHOTO;
capturelistener->surface_ = photoSurface;
photoSurface->RegisterConsumerListener((sptr<IBufferConsumerListener> &)capturelistener);

// Create the Photo Output
sptr<CaptureOutput> photoOutput = camManagerObj->CreatePhotoOutput(photoSurface);

// Add the output to the capture session
result = captureSession->AddOutput(photoOutput);

  16. Commit the configuration to Capture Session.

    result = captureSession->CommitConfig();

  17. Release the Camera Input that is removed from session.

    cameraInput->Release();

  18. Capture the Photo.

    result = ((sptr<PhotoOutput> &)photoOutput)->Capture();

  19. Release the Capture Session resources.

    captureSession->Release();

  20. Release the Camera Input to close the camera.

    cameraInput2->Release();

To configure flash mode

Flash mode can be configured on the Camera Input before capturing a photo or recording a video.

  1. Set the flash mode for photo capture:

    cameraInput->LockForControl();
cameraInput->SetFlashMode(OHOS_CAMERA_FLASH_MODE_OPEN);
cameraInput->UnlockForControl();

  2. Set the flash mode for video recording:

    cameraInput->LockForControl();
cameraInput->SetFlashMode(OHOS_CAMERA_FLASH_MODE_ALWAYS_OPEN);
cameraInput->UnlockForControl();

  3. Turnoff the flash:

    cameraInput->LockForControl();
cameraInput->SetFlashMode(OHOS_CAMERA_FLASH_MODE_CLOSE);
cameraInput->UnlockForControl();

Repositories Involved

multimedia_camera_framework