Advanced recognition settings
Be default elDoc IDP assumes that input files (images) are provided with average or good scanning quality, as such they are pre-processed with the default settings optimized for processing images of the average quality.
elDoc IDP also allows to process images provided with poor and bad quality (noisy background or color background, twisted and rotated photos of documents, etc.), but such scanned documents may require custom preprocessing settings. For this purpose elDoc IDP provides possibility to define custom parameters per RecoForm so when documents are to be processed with the given RecoForm they are at first place re-pre-processed with custom settings and only after that - IDP will be performed.
This page provides overview of the available advanced settings.
Recognition Form Advanced Settings
Advanced settings are set via the Advanced settings configuration area of the RecoForm.
Advanced settings description
Advanced setting param name | Data type | Default value | Description | Triggers custom pre-processing* |
---|---|---|---|---|
idp_recoform_table_ancundtblusebottom | Boolean | false | Use bottom left point of anchor for calculating location of table end | |
idp_recoform_use_source | Boolean | false | Switches RecoForm to use source image instead of the pre-processed image for the RecoForm processing | |
idp_recoform_custom_id | String | Id of the custom RecoForm which is loaded as plugin | ||
idp_recoform_morphology | String | Enables artifacts removal and sets size of the artifact to be removed during image pre-processing Min value: "0x0" (disabled) Max value: "99x99" | ||
idp_adaptive_threshold | Boolean | false | Enables adaptive threshold | |
idp_iterative_threshold | Boolean | false | Enables iterative threshold | |
idp_threshold_value | Integer | Defines custom thresholding value. When set to 0 - processing of this property will be ignored Min value: 0 (disabled) Max value: 255 | ||
idp_resolution_threshold | String | If source image resolution is below defined threshold in at least one dimension document will be sent to validation Min value: "10x10" Max value: "99999x99999" | ||
idp_ocr_mode_sparse_text | Boolean | false | Sets OCR recognition mode to Sparse Text (Find as much text as possible in no particular order) | |
idp_recoform_strict_layout | Boolean | false | Enforces strict template layout for the RecoForm. When this value set to true anchors' positions are used for calculating page scale | |
idp_ignore_text_layer | Boolean | false | Instructs system to ignore text-layer (if any available, relevant to the PDF files) and enforces document OCR | |
idp_recoform_keywords_classification | Boolean | false | Enables document classification based on the RecoForm keywords | |
idp_recoform_table_disable_row_merge | Boolean | false | Disables table rows merge | |
idp_recoform_table_enable_columns_detection | Boolean | false | Enables table columns width detection based on column border vertical lines (applicable for scan-images only) |
- "Triggers custom pre-processing" explanation:
System performs documents pre-processing on upload using most common pre-processing parameters. Due to specifics of some documents - generic parameters may not work best for such files, as such advanced settings are required to be defined on the RecoForms for such documents. Processing documents using such short-listed RecoForms with advanced settings marked with "triggers custom pre-processing" - causes system to perform pre-processing of the given document from scratch again, what may affect time required for the whole recognition queue processing.
Switch button "Force files preprocessing by elDoc" on the documents upload (Recognition queue) page enables the following advanced settings to be applied for document pre-processing: idp_ocr_mode_sparse_text, idp_iterative_threshold, idp_ignore_text_layer
Last modified: October 26, 2022