OCR
Tesseract - ReadText
Read text from an Image using a pre-trained OCR model. Note: Loading multiple languages will impact both speed and accuracy, as there is more work to do to decide on the applicable language, and there is more chance of hallucinating incorrect words.
IN
| Name | Type | Description |
|---|---|---|
| InImage | Image | The input Image containing the text to be recognized. |
| InLanguage | Enum < Language > | Language in which the text is written. |
| InModelPerformance | Enum < ModelPerformance > | Whether to use the Best or the Fast model available. |
| Name | String | Set the tool's name |
| Enable | Bool | Sets if current tool is enabled or not |
OUT
| Name | Type | Description |
|---|---|---|
| OutString | String | Recognized string of text. |
| Error | ErrorState | Gets the execution error message |
Tesseract - ReadTextByLevel
Read and segment text from an image using Tesseract pre-trained OCR model.
IN
| Name | Type | Description |
|---|---|---|
| InImage | Image | The input Image containing the text to be recognized. |
| InLanguage | Enum < Language > | Language in which the text is written. |
| InModelPerformance | Enum < ModelPerformance > | Whether to use the Best or the Fast model available. |
| InTextSplitLevel | Enum < TextSplitLevel > | Image segmentation mode. |
| Name | String | Set the tool's name |
| Enable | Bool | Sets if current tool is enabled or not |
OUT
| Name | Type | Description |
|---|---|---|
| OutString | StringArray | An array containing recognized string of text. |
| Error | ErrorState | Gets the execution error message |
Tesseract - ReadText_FromRegion
Read text from a Region using a pre-trained OCR model. Note: Loading multiple languages will impact both speed and accuracy, as there is more work to do to decide on the applicable language, and there is more chance of hallucinating incorrect words.
IN
| Name | Type | Description |
|---|---|---|
| InRegion | Region | The input Region containing the text to be recognized. |
| InLanguage | Enum < Language > | Language in which the text is written. |
| InModelPerformance | Enum < ModelPerformance > | Whether to use the Best or the Fast model available. |
| Name | String | Set the tool's name |
| Enable | Bool | Sets if current tool is enabled or not |
OUT
| Name | Type | Description |
|---|---|---|
| OutString | String | Recognized string of text. |
| Error | ErrorState | Gets the execution error message |
Paddle - FindText_Paddle
Find the region of the image containing text.
IN
| Name | Type | Description |
|---|---|---|
| InImage | Image | The input Image containing the text to be highlighted. |
| InModelPerformance | Enum < ModelPerformance > | Whether to use the Best or the Fast model available. |
| InThreshold | Float | Threshold applied to the model's score map to binarize and extract candidate text regions. |
| InBoxThreshold | Float | Minimum average score required for a detected box/region to be considered valid. |
| InUnclipRatio | Float | Ratio used to expand （'unclip'） the detected text region. A higher value increases the size of the output bounding box, useful for covering the full text area. |
| InInferenceDevice | Enum < InferenceDevice > | Device on which to run the model. |
| Name | String | Set the tool's name |
| Enable | Bool | Sets if current tool is enabled or not |
OUT
| Name | Type | Description |
|---|---|---|
| OutRectangle2D | Rectangle2DArray | An Array |
| Error | ErrorState | Gets the execution error message |
Paddle - ReadText_Paddle
Read and segment text from an image using Paddle pre-trained OCR model.
IN
| Name | Type | Description |
|---|---|---|
| InImage | Image | The input Image containing the text to be recognized. |
| InTextRegions | Rectangle2DArray | Array of Rectangle2D specifying regions to read. |
| InLanguage | Enum < Language > | Suggested language in which the text is written. The model can recognize multiple languages, but this can help it to focus on the right character set among the supported ones （e.g., English, Italian, Simplified Chinese, Traditional Chinese）. |
| InInferenceDevice | Enum < InferenceDevice > | Device on which to run the model. |
| Name | String | Set the tool's name |
| Enable | Bool | Sets if current tool is enabled or not |
OUT
| Name | Type | Description |
|---|---|---|
| OutString | StringArray | Recognized strings of text. |
| Error | ErrorState | Gets the execution error message |
EasyOCR - FindText_EasyOCR
Find the region of the image containing text.
IN
| Name | Type | Description |
|---|---|---|
| InImage | Image | The input Image containing the text to be highlighted. |
| InScoreThreshold | Float | CRAFT Model Score Map threshold |
| InAffinityThreshold | Float | CRAFT Model Affinity Map threshold |
| InInferenceDevice | Enum < InferenceDevice > | Device on which to run the model. |
| Name | String | Set the tool's name |
| Enable | Bool | Sets if current tool is enabled or not |
OUT
| Name | Type | Description |
|---|---|---|
| OutRectangle2D | Rectangle2DArray | An Array |
| Error | ErrorState | Gets the execution error message |
EasyOCR - ReadText_EasyOCR
Read and segment text from an image using EasyOCR pre-trained OCR model.
IN
| Name | Type | Description |
|---|---|---|
| InImage | Image | The input Image containing the text to be recognized. |
| InTextRegions | Rectangle2DArray | Array of Rectangle2D specifying regions to read. |
| InLanguage | Enum < Language > | Language in which the text is written. |
| InInferenceDevice | Enum < InferenceDevice > | Device on which to run the model. |
| Name | String | Set the tool's name |
| Enable | Bool | Sets if current tool is enabled or not |
OUT
| Name | Type | Description |
|---|---|---|
| OutString | StringArray | Recognized strings of text. |
| Error | ErrorState | Gets the execution error message |