### Install Dynamsoft Capture Vision Bundle

Source: https://github.com/dynamsoft/capture-vision-python-samples/blob/main/README.md

Use pip or pip3 to install the Dynamsoft Capture Vision bundle for Python.

```bash
pip install dynamsoft-capture-vision-bundle
```

```bash
pip3 install dynamsoft-capture-vision-bundle
```

--------------------------------

### Initialize License with Public Trial Key

Source: https://context7.com/dynamsoft/capture-vision-python-samples/llms.txt

Initialize the SDK license using a public trial key. This is a required step before any processing can occur. Ensure network access for the trial key. Handle potential errors during initialization.

```python
from dynamsoft_capture_vision_bundle import LicenseManager, EnumErrorCode

# Public trial key – requires internet access.
# Replace with your own key for production or offline use.
error_code, error_message = LicenseManager.init_license(
    "DLS2eyJvcmdhbml6YXRpb25JRCI6IjIwMDAwMSJ9"
)

if error_code not in (EnumErrorCode.EC_OK, EnumErrorCode.EC_LICENSE_WARNING):
    raise RuntimeError(
        f"License initialization failed – code: {error_code}, msg: {error_message}"
    )

# EC_LICENSE_WARNING is non-fatal (e.g. trial about to expire); log and continue.
if error_code == EnumErrorCode.EC_LICENSE_WARNING:
    print(f"License warning: {error_message}")
```

--------------------------------

### CaptureVisionRouter.init_settings_from_file

Source: https://context7.com/dynamsoft/capture-vision-python-samples/llms.txt

Loads a custom JSON template to configure the router. This allows for custom barcode formats, image-processing stages, and more. Must be called before `capture_multi_pages` if using a custom template name.

```APIDOC
## CaptureVisionRouter.init_settings_from_file – Load a custom JSON template

Replaces the router's built-in settings with a JSON template file that defines custom barcode formats, image-processing stages, semantic parsing tasks, and ROI definitions. Must be called before `capture_multi_pages` when a custom template name is used.

```python
from dynamsoft_capture_vision_bundle import CaptureVisionRouter, EnumErrorCode
from pathlib import Path

cvr = CaptureVisionRouter()

template_path = Path("CustomTemplates/ReadGS1AIBarcode.json")
error_code, error_msg = cvr.init_settings_from_file(str(template_path))

if error_code != EnumErrorCode.EC_OK:
    raise RuntimeError(f"Failed to load template: {error_msg}")

# The router now recognises the "ReadGS1AIBarcode" template name.
```
```

--------------------------------

### Initialize License for Dynamsoft Capture Vision

Source: https://github.com/dynamsoft/capture-vision-python-samples/blob/main/README.md

Initialize the Dynamsoft Capture Vision SDK with a license key. A free public trial license requires network connection, while a requested trial license works offline.

```python
from dcv_sdk.core.license_manager import LicenseManager

# Use a free public trial license (requires network connection)
LicenseManager.init_license("DLS20220101")

# Or use a 30-day free trial license (works offline)
# LicenseManager.init_license("YOUR_30_DAY_TRIAL_LICENSE")
```

--------------------------------

### LicenseManager.init_license

Source: https://context7.com/dynamsoft/capture-vision-python-samples/llms.txt

Initializes the DCV runtime license. This must be called before creating a CaptureVisionRouter. A free trial key is provided, but a production key is recommended for long-term use.

```APIDOC
## LicenseManager.init_license – Activate the SDK with a license key

Initializes the DCV runtime license before any processing can occur. The call must succeed (or return only a warning) before a `CaptureVisionRouter` is created. A free public trial key that requires network access is bundled with every sample; offline 30-day keys are available from the Dynamsoft portal.

```python
from dynamsoft_capture_vision_bundle import LicenseManager, EnumErrorCode

# Public trial key – requires internet access.
# Replace with your own key for production or offline use.
error_code, error_message = LicenseManager.init_license(
    "DLS2eyJvcmdhbml6YXRpb25JRCI6IjIwMDAwMSJ9"
)

if error_code not in (EnumErrorCode.EC_OK, EnumErrorCode.EC_LICENSE_WARNING):
    raise RuntimeError(
        f"License initialization failed – code: {error_code}, msg: {error_message}"
    )

# EC_LICENSE_WARNING is non-fatal (e.g. trial about to expire); log and continue.
if error_code == EnumErrorCode.EC_LICENSE_WARNING:
    print(f"License warning: {error_message}")
```
```

--------------------------------

### Create CaptureVisionRouter Instance

Source: https://context7.com/dynamsoft/capture-vision-python-samples/llms.txt

Instantiate the `CaptureVisionRouter`, which serves as the central engine for all image analysis tasks. No further configuration is needed for built-in templates.

```python
from dynamsoft_capture_vision_bundle import CaptureVisionRouter

cvr = CaptureVisionRouter()
# The router is now ready – no further configuration is required for built-in templates.
```

--------------------------------

### Load Custom Template from JSON File

Source: https://context7.com/dynamsoft/capture-vision-python-samples/llms.txt

Load custom settings from a JSON template file into the `CaptureVisionRouter`. This allows for custom barcode formats, image processing stages, and more. Must be called before using a custom template name in `capture_multi_pages`.

```python
from dynamsoft_capture_vision_bundle import CaptureVisionRouter, EnumErrorCode
from pathlib import Path

cvr = CaptureVisionRouter()

template_path = Path("CustomTemplates/ReadGS1AIBarcode.json")
error_code, error_msg = cvr.init_settings_from_file(str(template_path))

if error_code != EnumErrorCode.EC_OK:
    raise RuntimeError(f"Failed to load template: {error_msg}")

# The router now recognises the "ReadGS1AIBarcode" template name.
```

--------------------------------

### Detect and Normalize Document Image

Source: https://context7.com/dynamsoft/capture-vision-python-samples/llms.txt

Uses the PT_DETECT_AND_NORMALIZE_DOCUMENT preset to find, deskew, and save a normalized document image. Ensure the license is initialized before use.

```python
from dynamsoft_capture_vision_bundle import (
    CaptureVisionRouter, LicenseManager, EnumErrorCode,
    EnumPresetTemplate, FileImageTag, ImageIO
)

LicenseManager.init_license("DLS2eyJvcmdhbml6YXRpb25JRCI6IjIwMDAwMSJ9")
cvr = CaptureVisionRouter()

result_array = cvr.capture_multi_pages(
    "Images/document-sample.jpg",
    EnumPresetTemplate.PT_DETECT_AND_NORMALIZE_DOCUMENT
)

image_io = ImageIO()

for i, result in enumerate(result_array.get_results()):
    tag = result.get_original_image_tag()
    page_number = tag.get_page_number() + 1 if isinstance(tag, FileImageTag) else i + 1

    if result.get_error_code() not in (EnumErrorCode.EC_OK,
                                        EnumErrorCode.EC_UNSUPPORTED_JSON_KEY_WARNING):
        print(f"Page {page_number} error:", result.get_error_string())
        continue

    processed = result.get_processed_document_result()
    if processed is None or not processed.get_enhanced_image_result_items():
        print(f"Page {page_number}: no document found")
        continue

    for idx, item in enumerate(processed.get_enhanced_image_result_items()):
        out_path = f"Page_{page_number}_enhanced_{idx}.png"
        image_data = item.get_image_data()
        if image_data:
            err, msg = image_io.save_to_file(image_data, out_path)
            if err == 0:
                print(f"Saved: {out_path}")
            else:
                print(f"Save failed: {err} – {msg}")

# Expected output:
# Saved: Page_1_enhanced_0.png
```

--------------------------------

### Process Image File with CaptureVisionRouter

Source: https://context7.com/dynamsoft/capture-vision-python-samples/llms.txt

Process an image or PDF file using the `CaptureVisionRouter`. This function runs the entire DCV pipeline and returns a `CapturedResultArray` containing per-page results.

```python
from dynamsoft_capture_vision_bundle import (
    CaptureVisionRouter, LicenseManager, EnumErrorCode,
    EnumPresetTemplate, FileImageTag
)

LicenseManager.init_license("DLS2eyJvcmdhbml6YXRpb25JRCI6IjIwMDAwMSJ9")
cvr = CaptureVisionRouter()


```

--------------------------------

### Save ImageData to File using ImageIO

Source: https://context7.com/dynamsoft/capture-vision-python-samples/llms.txt

Persists an ImageData object to a file. The output format is determined by the file extension (e.g., PNG, JPEG, BMP). Ensure the output path is correctly specified.

```python
from dynamsoft_capture_vision_bundle import (
    CaptureVisionRouter, LicenseManager, EnumErrorCode,
    EnumPresetTemplate, ImageIO
)

LicenseManager.init_license("DLS2eyJvcmdhbml6YXRpb25JRCI6IjIwMDAwMSJ9")
cvr = CaptureVisionRouter()
image_io = ImageIO()

result_array = cvr.capture_multi_pages(
    "Images/document-sample.jpg",
    EnumPresetTemplate.PT_DETECT_AND_NORMALIZE_DOCUMENT
)

for i, result in enumerate(result_array.get_results()):
    processed = result.get_processed_document_result()
    if processed is None:
        continue
    for j, item in enumerate(processed.get_enhanced_image_result_items()):
        image_data = item.get_image_data()
        if image_data is None:
            continue
        out_path = f"enhanced_doc_p{i+1}_{j}.png"   # PNG, JPEG, BMP all supported
        error_code, error_msg = image_io.save_to_file(image_data, out_path)
        if error_code == 0:
            print(f"Saved {out_path}")
        else:
            print(f"Save failed ({error_code}): {error_msg}")
```

--------------------------------

### CaptureVisionRouter.capture_multi_pages

Source: https://context7.com/dynamsoft/capture-vision-python-samples/llms.txt

Processes an image or PDF file through the full DCV pipeline. It returns a `CapturedResultArray` containing results for each page.

```APIDOC
## CaptureVisionRouter.capture_multi_pages – Process an image file

Runs the full DCV pipeline (region pre-detection → localization → decoding/recognition → semantic parsing) on every page of the supplied image or PDF. Returns a `CapturedResultArray`; iterate its `.get_results()` list to inspect per-page output.

```python
from dynamsoft_capture_vision_bundle import (
    CaptureVisionRouter, LicenseManager, EnumErrorCode,
    EnumPresetTemplate, FileImageTag
)

LicenseManager.init_license("DLS2eyJvcmdhbml6YXRpb25JRCI6IjIwMDAwMSJ9")
cvr = CaptureVisionRouter()


```

--------------------------------

### Read GS1 AI Barcodes from Image

Source: https://context7.com/dynamsoft/capture-vision-python-samples/llms.txt

Reads GS1 AI barcodes from an image file and prints the parsed AI, code, and value. Ensure the 'ReadGS1AIBarcode' template is configured.

```python
result_array = cvr.capture_multi_pages(
    "Images/gs1-ai-sample.png",
    "ReadGS1AIBarcode"
)

for i, result in enumerate(result_array.get_results()):
    if result.get_error_code() not in (EnumErrorCode.EC_OK,
                                        EnumErrorCode.EC_UNSUPPORTED_JSON_KEY_WARNING):
        print("Error:", result.get_error_string())
        continue

    parsed = result.get_raw_passed_result().get_parsed_result()
    if not parsed or not parsed.get_items():
        print(f"Page {i+1}: no GS1 AIs found")
        continue

    for item in parsed.get_items():
        for key, ai_code, value in parse_gs1_result(item):
            print(f"AI: {key} ({ai_code}), Value: {value}")
```

--------------------------------

### CaptureVisionRouter

Source: https://context7.com/dynamsoft/capture-vision-python-samples/llms.txt

The central processing engine for all image-analysis tasks. A single instance can be reused across multiple images and template names.

```APIDOC
## CaptureVisionRouter – Central processing engine

`CaptureVisionRouter` is the single entry point for all image-analysis tasks. After construction it is ready to use with any built-in template. One router instance can be reused across multiple images and multiple template names.

```python
from dynamsoft_capture_vision_bundle import CaptureVisionRouter

cvr = CaptureVisionRouter()
# The router is now ready – no further configuration is required for built-in templates.
```
```

--------------------------------

### ImageIO.save_to_file

Source: https://context7.com/dynamsoft/capture-vision-python-samples/llms.txt

Persists an ImageData object to disk. The output format is inferred from the file extension. Supported formats include PNG, JPEG, and BMP.

```APIDOC
## ImageIO.save_to_file

### Description
Persists an `ImageData` object to disk. The output format is inferred from the file extension.

### Method
`save_to_file(image_data: ImageData, out_path: str) -> Tuple[int, str]`

### Parameters
#### Path Parameters
- **image_data** (ImageData) - Required - The image data object to save.
- **out_path** (str) - Required - The file path to save the image to. The extension determines the format (e.g., .png, .jpg, .bmp).

### Return Value
A tuple containing:
- **error_code** (int) - 0 if successful, non-zero otherwise.
- **error_msg** (str) - A message describing the error if an error occurred.
```

--------------------------------

### Read Passport and ID Information

Source: https://context7.com/dynamsoft/capture-vision-python-samples/llms.txt

Processes passport and ID images to extract information. Ensure the 'ReadPassportAndId' template is available. Handles potential errors during parsing and saves a portrait crop for passports.

```python
LicenseManager.init_license("DLS2eyJvcmdhbml6YXRpb25JRCI6IjIwMDAwMSJ9")
cvr = CaptureVisionRouter()

irr = MyIRReceiver()
irm = cvr.get_intermediate_result_manager()
irm.add_result_receiver(irr)

result_array = cvr.capture_multi_pages(
    "Images/passport-sample.jpg",
    "ReadPassportAndId"
)

image_io = ImageIO()
img_proc = ImageProcessor()

for i, result in enumerate(result_array.get_results()):
    if result.get_error_code() not in (EnumErrorCode.EC_OK,
                                        EnumErrorCode.EC_UNSUPPORTED_JSON_KEY_WARNING):
        print("Error:", result.get_error_string())
        continue

    parsed = result.get_parsed_result()
    if not parsed or not parsed.get_items():
        print("No MRZ found.")
        continue

    hash_id = result.get_original_image_hash_id()
    is_passport = False

    for item in parsed.get_items():
        code_type = item.get_code_type()
        is_passport = (code_type == "MRTD_TD3_PASSPORT")

        def field(name):
            v = item.get_field_value(name)
            s = item.get_field_validation_status(name)
            return v if v and s != EnumValidationStatus.VS_FAILED else None

        print(f"Type:        {code_type}")
        print(f"Surname:     {field('primaryIdentifier')}")
        print(f"Given name:  {field('secondaryIdentifier')}")
        print(f"Nationality: {field('nationality')}")
        print(f"DOB:         {field('dateOfBirth')}")
        print(f"Expiry:      {field('dateOfExpiry')}")
        print(f"Line 1:      {field('line1')}")
        print(f"Line 2:      {field('line2')}")

    # Save portrait crop for passports
    if is_passport:
        original_image = next(
            (it.get_image_data() for it in result.get_items()
             if isinstance(it, OriginalImageResultItem)), None
        )
        quad = irr.get_portrait_quad(hash_id)
        if quad and original_image:
            ret, portrait = img_proc.crop_and_deskew_image(original_image, quad)
            if ret == EnumErrorCode.EC_OK:
                err, msg = image_io.save_to_file(portrait, f"portrait_page{i+1}.png")
                print(f"Portrait saved: portrait_page{i+1}.png" if err == 0 else f"Save failed: {msg}")
```

--------------------------------

### Define Custom Capture Pipeline with JSON Template

Source: https://context7.com/dynamsoft/capture-vision-python-samples/llms.txt

Defines a custom capture pipeline named 'ReadGS1AIBarcode' using a JSON template. This template configures various task settings for barcode reading and semantic processing.

```json
{
  "CaptureVisionTemplates": [
    {
      "Name": "ReadGS1AIBarcode",
      "ImageROIProcessingNameArray": ["roi_gs1_ai_barcode"],
      "SemanticProcessingNameArray": ["sp_gs1_ai"]
    }
  ],
  "BarcodeReaderTaskSettingOptions": [
    {
      "Name": "task_gs1_ai_barcode",
      "ExpectedBarcodesCount": 1,
      "BarcodeFormatIds": ["BF_GS1_DATABAR", "BF_CODE_128", "BF_GS1_COMPOSITE"],
      "BarcodeFormatSpecificationNameArray": ["bfs_gs1_ai"]
    }
  ],
  "BarcodeFormatSpecificationOptions": [
    {
      "Name": "bfs_gs1_ai",
      "BarcodeFormatIds": ["BF_GS1_DATABAR", "BF_CODE_128", "BF_GS1_COMPOSITE"],
      "IncludeTrailingCheckDigit": 0,
      "IncludeImpliedAI01": 1
    }
  ],
  "CodeParserTaskSettingOptions": [
    {
      "Name": "dcp_gs1_ai",
      "CodeSpecifications": ["GS1_AI"]
    }
  ],
  "SemanticProcessingOptions": [
    {
      "Name": "sp_gs1_ai",
      "ReferenceObjectFilter": {
        "ReferenceTargetROIDefNameArray": ["roi_gs1_ai_barcode"]
      },
      "TaskSettingNameArray": ["dcp_gs1_ai"]
    }
  ],
  "TargetROIDefOptions": [
    {
      "Name": "roi_gs1_ai_barcode",
      "TaskSettingNameArray": ["task_gs1_ai_barcode"]
    }
  ]
}
```

--------------------------------

### MRZ Scanner with Intermediate Result Receiver

Source: https://context7.com/dynamsoft/capture-vision-python-samples/llms.txt

This snippet demonstrates how to use the `ReadPassportAndId` preset to scan MRZ zones on travel documents. It also shows the `IntermediateResultReceiver` pattern for capturing and processing intermediate frames, which is crucial for tasks like portrait extraction.

```python
from dynamsoft_capture_vision_bundle import (
    CaptureVisionRouter, LicenseManager, EnumErrorCode,
    IntermediateResultReceiver, IntermediateResultExtraInfo,
    DeskewedImageUnit, LocalizedTextLinesUnit, ScaledColourImageUnit,
    RecognizedTextLinesUnit, DetectedQuadsUnit,
    ParsedResultItem, EnumValidationStatus, FileImageTag,
    IdentityProcessor, ImageProcessor, ImageIO, Quadrilateral,
    OriginalImageResultItem
)

# ── Intermediate result collection ──────────────────────────────────────────

class MyIRReceiver(IntermediateResultReceiver):
    def __init__(self):
        super().__init__()
        self._groups = {}   # hash_id → dict of units

    def _group(self, hash_id):
        if hash_id not in self._groups:
            self._groups[hash_id] = {}
        return self._groups[hash_id]

    def on_deskewed_image_received(self, result: DeskewedImageUnit,
                                    info: IntermediateResultExtraInfo):
        if info.is_section_level_result:
            self._group(result.get_original_image_hash_id())["deskewed"] = result

    def on_localized_text_lines_received(self, result: LocalizedTextLinesUnit,
                                          info: IntermediateResultExtraInfo):
        if info.is_section_level_result:
            self._group(result.get_original_image_hash_id())["localized_lines"] = result

    def on_scaled_colour_image_unit_received(self, result: ScaledColourImageUnit,
                                              info: IntermediateResultExtraInfo):
        self._group(result.get_original_image_hash_id())["scaled_colour"] = result

    def on_recognized_text_lines_received(self, result: RecognizedTextLinesUnit,
                                           info: IntermediateResultExtraInfo):
        if info.is_section_level_result:
            self._group(result.get_original_image_hash_id())["recognized_lines"] = result

    def on_detected_quads_received(self, result: DetectedQuadsUnit,
                                    info: IntermediateResultExtraInfo):
        if info.is_section_level_result:
            self._group(result.get_original_image_hash_id())["quads"] = result

    def get_portrait_quad(self, hash_id: str):
        g = self._groups.get(hash_id, {})
        id_proc = IdentityProcessor()
        ret, quad = id_proc.find_portrait_zone(
            g.get("scaled_colour"), g.get("localized_lines"),
            g.get("recognized_lines"), g.get("quads"), g.get("deskewed")
        )
        return quad if ret == EnumErrorCode.EC_OK else None

```

--------------------------------

### Parse Driver License Barcodes

Source: https://context7.com/dynamsoft/capture-vision-python-samples/llms.txt

Uses the "ReadDriversLicense" preset to decode AAMVA barcodes from driver licenses. It extracts and validates fields using helper functions. Ensure the license is initialized.

```python
from dynamsoft_capture_vision_bundle import (
    CaptureVisionRouter, LicenseManager, EnumErrorCode,
    EnumValidationStatus, ParsedResultItem, FileImageTag
)

LicenseManager.init_license("DLS2eyJvcmdhbml6YXRpb25JRCI6IjIwMDAwMSJ9")
cvr = CaptureVisionRouter()

result_array = cvr.capture_multi_pages(
    "Images/driver-license-sample.jpg",
    "ReadDriversLicense"
)

def extract_dl_field(item: ParsedResultItem, *field_names):
    """Return the first validated field value from a list of candidate names."""
    for name in field_names:
        value = item.get_field_value(name)
        status = item.get_field_validation_status(name)
        if value is not None and status != EnumValidationStatus.VS_FAILED:
            return value
    return None

for i, result in enumerate(result_array.get_results()):
    if result.get_error_code() not in (EnumErrorCode.EC_OK,
                                        EnumErrorCode.EC_UNSUPPORTED_JSON_KEY_WARNING):
        print("Error:", result.get_error_string())
        continue

    parsed = result.get_parsed_result()
    if parsed is None or not parsed.get_items():
        print("No driver license found.")
        continue

    for item in parsed.get_items():
        license_number   = extract_dl_field(item, "licenseNumber")
        last_name        = extract_dl_field(item, "lastName", "surName")
        given_name       = extract_dl_field(item, "givenName")
        gender           = extract_dl_field(item, "sex", "gender")
        birth_date       = extract_dl_field(item, "birthDate")
        expiration_date  = extract_dl_field(item, "expirationDate")
        vehicle_class    = extract_dl_field(item, "vehicleClass")

        print(f"License #:   {license_number}")
        print(f"Name:        {last_name}, {given_name}")
        print(f"Gender:      {gender}")
        print(f"DOB:         {birth_date}")
        print(f"Expires:     {expiration_date}")
        print(f"Class:       {vehicle_class}")

# Expected output:
# License #:   S99900589
# Name:        SAMPLE, ALEXANDER
```

--------------------------------

### Parse GS1 AI Barcode Data

Source: https://context7.com/dynamsoft/capture-vision-python-samples/llms.txt

Parses GS1 Application Identifiers from barcodes using a custom JSON template. Requires 'ReadGS1AIBarcode.json' in 'CustomTemplates/'. Helper functions group fields and identify AI tokens.

```python
import re
from collections import defaultdict
from pathlib import Path
from dynamsoft_capture_vision_bundle import (
    CaptureVisionRouter, LicenseManager, EnumErrorCode,
    ParsedResultItem, FileImageTag
)

def group_by_first_layer(names):
    grouped = defaultdict(list)
    for name in names:
        grouped[name.split('.')[0]].append(name)
    return grouped

def is_ai(token):
    return re.fullmatch(r"\d+|(\d+n)", token) is not None

def parse_gs1_result(item: ParsedResultItem):
    grouped = group_by_first_layer(item.get_all_field_names())
    entries = []
    for key, fields in grouped.items():
        if not is_ai(key):
            continue
        ai = data = None
        for f in fields:
            last = f.split('.')[-1]
            if last == key + 'AI':
                ai = item.get_field_value(f)
            elif last == key + 'Data':
                data = item.get_field_value(f)
        entries.append((key, ai, data))
    return entries

# ── Setup ────────────────────────────────────────────────────────────────────

LicenseManager.init_license("DLS2eyJvcmdhbml6YXRpb25JRCI6IjIwMDAwMSJ9")
cvr = CaptureVisionRouter()

template_path = Path("CustomTemplates/ReadGS1AIBarcode.json")
err, msg = cvr.init_settings_from_file(str(template_path))
if err != EnumErrorCode.EC_OK:
    raise RuntimeError(f"Template load failed: {msg}")

```

--------------------------------

### Intercept Intermediate Results with Custom Receiver

Source: https://context7.com/dynamsoft/capture-vision-python-samples/llms.txt

Subclass `IntermediateResultReceiver` to capture and process intermediate image frames and geometry data during the capture pipeline. This is useful for debugging or advanced image analysis.

```python
from dynamsoft_capture_vision_bundle import (
    CaptureVisionRouter, IntermediateResultReceiver,
    IntermediateResultExtraInfo, ScaledColourImageUnit,
    LicenseManager, EnumErrorCode
)

class DebugReceiver(IntermediateResultReceiver):
    def __init__(self):
        super().__init__()
        self.frames = []

    def on_scaled_colour_image_unit_received(
        self,
        result: ScaledColourImageUnit,
        info: IntermediateResultExtraInfo
    ):
        img = result.get_image_data()
        self.frames.append(img)
        print(f"Captured intermediate frame: {img.get_width()}×{img.get_height()} px")

LicenseManager.init_license("DLS2eyJvcmdhbml6YXRpb25JRCI6IjIwMDAwMSJ9")
cvr = CaptureVisionRouter()

receiver = DebugReceiver()
irm = cvr.get_intermediate_result_manager()
irm.add_result_receiver(receiver)

cvr.capture_multi_pages("Images/passport-sample.jpg", "ReadPassportAndId")

print(f"Total intermediate frames captured: {len(receiver.frames)}")
# Remove receiver when no longer needed
irm.remove_result_receiver(receiver)
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.