CtrlK
BlogDocsLog inGet started
Tessl Logo

tessl/maven-org-bytedeco--tesseract-platform

JavaCPP platform aggregator for Tesseract OCR native libraries providing cross-platform OCR capabilities in Java applications

—

Pending
Overview
Eval results
Files

configuration.mddocs/

Configuration

Comprehensive configuration system with hundreds of parameters controlling OCR behavior, page segmentation, character recognition, and output formatting. Tesseract provides fine-grained control over recognition algorithms and processing options.

Capabilities

Parameter Management

Set and retrieve configuration parameters that control Tesseract's behavior during initialization and recognition.

public class TessBaseAPI {
    // Parameter setting
    public boolean SetVariable(String name, String value);
    public boolean SetDebugVariable(String name, String value);
    
    // Parameter retrieval
    public boolean GetIntVariable(String name, int[] value);
    public boolean GetBoolVariable(String name, boolean[] value);
    public boolean GetDoubleVariable(String name, double[] value);
    public String GetStringVariable(String name);
    
    // Generic parameter access
    public native @Cast("bool") boolean GetVariableAsString(String name, @StdString @Cast({"char*", "std::string*"}) BytePointer val);
    
    // Parameter introspection
    public void PrintVariables(Pointer fp);
}

Parameter Types:

  • String Parameters: Language codes, file paths, character sets
  • Integer Parameters: Thresholds, limits, numeric options
  • Boolean Parameters: Feature toggles, mode switches
  • Double Parameters: Floating-point thresholds, scaling factors

Usage Example

TessBaseAPI api = new TessBaseAPI();
api.Init(null, "eng");

// Set character blacklist (ignore these characters)
api.SetVariable("tessedit_char_blacklist", "xyz@#$");

// Enable numeric-only mode
api.SetVariable("classify_bln_numeric_mode", "1");

// Set minimum word length
api.SetVariable("textord_min_linesize", "2.5");

// Enable debug output
api.SetDebugVariable("textord_debug_tabfind", "1");

// Check if parameter exists and get value
int[] min_chars = new int[1];
if (api.GetIntVariable("textord_min_chars_in_word", min_chars)) {
    System.out.println("Minimum characters per word: " + min_chars[0]);
}

// Get string parameter
String lang = api.GetStringVariable("tessedit_init_config_only");
System.out.println("Config only mode: " + lang);

OCR Engine Mode Configuration

Control which OCR engine and neural network models are used for recognition.

// OCR Engine Mode constants
public static final int OEM_TESSERACT_ONLY = 0;           // Legacy Tesseract (deprecated)
public static final int OEM_LSTM_ONLY = 1;                // LSTM neural network only  
public static final int OEM_TESSERACT_LSTM_COMBINED = 2;  // Combined legacy + LSTM (deprecated)
public static final int OEM_DEFAULT = 3;                  // Default (currently LSTM only)

Usage Example

TessBaseAPI api = new TessBaseAPI();

// Use LSTM-only engine (recommended)
int result = api.Init(null, "eng", OEM_LSTM_ONLY);

// Or use default engine mode
int result2 = api.Init(null, "eng", OEM_DEFAULT);

// Check which engine is active
String engine_mode = api.GetStringVariable("tessedit_ocr_engine_mode");
System.out.println("Active OCR engine: " + engine_mode);

Page Segmentation Configuration

Configure how Tesseract analyzes page layout and identifies text regions.

public class TessBaseAPI {
    // Page segmentation mode control
    public void SetPageSegMode(int mode);
    public int GetPageSegMode();
}

// Page segmentation mode constants (detailed)
public static final int PSM_OSD_ONLY = 0;              // Orientation/script detection only
public static final int PSM_AUTO_OSD = 1;              // Automatic with OSD
public static final int PSM_AUTO_ONLY = 2;             // Automatic without OSD
public static final int PSM_AUTO = 3;                  // Fully automatic (default)
public static final int PSM_SINGLE_COLUMN = 4;         // Single column text
public static final int PSM_SINGLE_BLOCK_VERT_TEXT = 5; // Single vertical block
public static final int PSM_SINGLE_BLOCK = 6;          // Single uniform block
public static final int PSM_SINGLE_LINE = 7;           // Single text line
public static final int PSM_SINGLE_WORD = 8;           // Single word
public static final int PSM_CIRCLE_WORD = 9;           // Single word in circle
public static final int PSM_SINGLE_CHAR = 10;          // Single character
public static final int PSM_SPARSE_TEXT = 11;          // Sparse text find
public static final int PSM_SPARSE_TEXT_OSD = 12;      // Sparse text with OSD
public static final int PSM_RAW_LINE = 13;             // Raw line (bypass hacks)

Usage Example

TessBaseAPI api = new TessBaseAPI();
api.Init(null, "eng");

// Configure for different document types
switch (documentType) {
    case NEWSPAPER:
        api.SetPageSegMode(PSM_AUTO);  // Multi-column with auto detection
        break;
        
    case BUSINESS_CARD:
        api.SetPageSegMode(PSM_SINGLE_BLOCK);  // Simple layout
        break;
        
    case LICENSE_PLATE:
        api.SetPageSegMode(PSM_SINGLE_LINE);   // Single line of text
        break;
        
    case CAPTCHA:
        api.SetPageSegMode(PSM_SINGLE_WORD);   // Single word
        break;
        
    case RECEIPT:
        api.SetPageSegMode(PSM_SINGLE_COLUMN); // Vertical list
        break;
}

// Verify current mode
int currentMode = api.GetPageSegMode();
System.out.println("Page segmentation mode: " + currentMode);

Character Recognition Configuration

Fine-tune character recognition algorithms and thresholds.

Common Character Recognition Parameters

// Character filtering and validation
api.SetVariable("tessedit_char_whitelist", "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz ");
api.SetVariable("tessedit_char_blacklist", "!@#$%^&*()");

// Numeric mode for numbers-only recognition
api.SetVariable("classify_bln_numeric_mode", "1");

// Case sensitivity
api.SetVariable("unicharset_use_script_ider", "1");

// Character segmentation
api.SetVariable("chop_enable", "1");              // Enable character chopping
api.SetVariable("use_new_state_cost", "1");       // Use improved state cost
api.SetVariable("segment_segcost_rating", "1");   // Enable segmentation cost rating

// Word recognition
api.SetVariable("save_best_choices", "1");        // Save alternative choices
api.SetVariable("language_model_penalty_non_dict_word", "0.15");
api.SetVariable("language_model_penalty_non_freq_dict_word", "0.1");

Usage Example

TessBaseAPI api = new TessBaseAPI();
api.Init(null, "eng");

// Configure for license plate recognition
api.SetPageSegMode(PSM_SINGLE_LINE);
api.SetVariable("tessedit_char_whitelist", "ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789");
api.SetVariable("classify_bln_numeric_mode", "0");  // Allow letters
api.SetVariable("textord_min_chars_in_word", "3");

// Configure for invoice numbers
api.SetPageSegMode(PSM_SINGLE_WORD);
api.SetVariable("tessedit_char_whitelist", "0123456789-");
api.SetVariable("classify_bln_numeric_mode", "1");

// Configure for general document with high accuracy
api.SetPageSegMode(PSM_AUTO);
api.SetVariable("tessedit_pageseg_mode", "3");
api.SetVariable("classify_enable_learning", "1");
api.SetVariable("classify_enable_adaptive_matcher", "1");

Image Processing Configuration

Control image preprocessing and enhancement algorithms.

Image Processing Parameters

// Image enhancement
api.SetVariable("textord_heavy_nr", "1");           // Enable heavy noise reduction
api.SetVariable("textord_noise_rejrows", "1");      // Reject noisy rows
api.SetVariable("textord_noise_rejwords", "1");     // Reject noisy words

// Skew detection and correction
api.SetVariable("textord_straight_baselines", "0"); // Allow curved baselines
api.SetVariable("textord_old_baselines", "1");      // Use old baseline fitting
api.SetVariable("textord_debug_baselines", "0");    // Debug baseline detection

// Text line finding
api.SetVariable("textord_really_old_xheight", "0"); // Use new x-height algorithm
api.SetVariable("textord_fix_xheight_bug", "1");    // Fix x-height bugs
api.SetVariable("textord_fix_makerow_bug", "1");    // Fix row-making bugs

// Edge detection
api.SetVariable("edges_use_new_outline_complexity", "1");
api.SetVariable("edges_max_children_per_outline", "10");

Usage Example

TessBaseAPI api = new TessBaseAPI();
api.Init(null, "eng");

// Configure for poor quality scanned documents
api.SetVariable("textord_heavy_nr", "1");
api.SetVariable("textord_noise_rejrows", "1");
api.SetVariable("textord_noise_rejwords", "1");
api.SetVariable("edges_use_new_outline_complexity", "1");

// Configure for high-quality printed text
api.SetVariable("textord_really_old_xheight", "0");
api.SetVariable("textord_fix_xheight_bug", "1");
api.SetVariable("classify_enable_adaptive_matcher", "1");

// Configure for handwritten text
api.SetVariable("classify_enable_learning", "1");
api.SetVariable("classify_adapt_proto_threshold", "230");
api.SetVariable("classify_adapt_feature_threshold", "230");

Debug and Logging Configuration

Enable detailed logging and debug output for troubleshooting and analysis.

Debug Parameters

// General debug output
api.SetDebugVariable("classify_debug_level", "2");
api.SetDebugVariable("textord_debug_tabfind", "1");
api.SetDebugVariable("textord_show_initial_words", "1");

// Image processing debug
api.SetDebugVariable("textord_debug_images", "1");
api.SetDebugVariable("textord_debug_to_screen", "1");

// Classification debug
api.SetDebugVariable("matcher_debug_level", "2");
api.SetDebugVariable("stopper_debug_level", "1");

// Layout analysis debug
api.SetDebugVariable("equationdetect_save_bi_image", "1");
api.SetDebugVariable("paragraph_debug_level", "1");

Usage Example

TessBaseAPI api = new TessBaseAPI();
api.Init(null, "eng");

// Enable comprehensive debugging for development
if (debugMode) {
    api.SetDebugVariable("classify_debug_level", "3");
    api.SetDebugVariable("textord_debug_tabfind", "1");
    api.SetDebugVariable("textord_debug_images", "1");
    api.SetDebugVariable("matcher_debug_level", "2");
    
    // Set debug output directory
    api.SetOutputName("/tmp/tesseract_debug");
}

// Enable specific debugging for problem analysis
if (layoutProblems) {
    api.SetDebugVariable("textord_show_initial_words", "1");
    api.SetDebugVariable("paragraph_debug_level", "2");
}

if (characterRecognitionProblems) {
    api.SetDebugVariable("classify_debug_level", "2");
    api.SetDebugVariable("stopper_debug_level", "1");
}

Performance and Resource Configuration

Control memory usage, processing speed, and computational resources.

Performance Parameters

// Processing timeouts
api.SetVariable("tessedit_timeout_millisecs", "30000");  // 30 second timeout

// Memory management
api.SetVariable("max_permuter_attempts", "10000");       // Limit word permutations
api.SetVariable("stopper_smallword_size", "2");          // Small word threshold

// Processing limits
api.SetVariable("language_model_penalty_increment", "0.01");
api.SetVariable("segment_penalty_dict_nonword", "1.25");
api.SetVariable("segment_penalty_garbage", "1.50");

// Quality vs speed trade-offs
api.SetVariable("tessedit_resegment_from_boxes", "1");   // Re-segment from boxes
api.SetVariable("tessedit_resegment_from_line_boxes", "1");
api.SetVariable("tessedit_train_from_boxes", "0");       // Disable training mode

Usage Example

TessBaseAPI api = new TessBaseAPI();
api.Init(null, "eng");

// Configure for high-speed processing
api.SetVariable("tessedit_timeout_millisecs", "5000");   // 5 second limit
api.SetVariable("max_permuter_attempts", "5000");        // Reduce attempts
api.SetVariable("stopper_smallword_size", "3");          // Larger small word threshold

// Configure for high-accuracy processing
api.SetVariable("tessedit_timeout_millisecs", "60000");  // 60 second limit
api.SetVariable("max_permuter_attempts", "20000");       // More attempts
api.SetVariable("classify_enable_adaptive_matcher", "1");
api.SetVariable("classify_enable_learning", "1");

// Configure for batch processing
api.SetVariable("tessedit_do_invert", "1");              // Auto-invert if needed
api.SetVariable("tessedit_write_images", "0");           // Don't save debug images
api.SetVariable("tessedit_dump_pageseg_images", "0");    // Don't save segmentation images

Configuration File Support

Load configuration from files and manage configuration sets.

// Configuration file parameters
api.SetVariable("tessedit_init_config_only", "0");       // Load config files
api.SetVariable("user_defined_dpi", "300");              // Set DPI if not detected

Usage Example

// Create configuration files for different use cases

// config_highaccuracy.txt
"""
classify_enable_adaptive_matcher 1
classify_enable_learning 1
tessedit_timeout_millisecs 60000
textord_heavy_nr 1
max_permuter_attempts 20000
"""

// config_speed.txt  
"""
tessedit_timeout_millisecs 5000
max_permuter_attempts 5000
classify_enable_adaptive_matcher 0
classify_enable_learning 0
"""

// Use configuration files
TessBaseAPI api = new TessBaseAPI();
// Configuration files are loaded automatically from tessdata/configs/
// when specified in Init
api.Init("/usr/share/tessdata", "eng+config_highaccuracy");

Common Configuration Patterns

Document Type Specific Configurations

public class TesseractConfigurations {
    
    public static void configureForNewspaper(TessBaseAPI api) {
        api.SetPageSegMode(PSM_AUTO);
        api.SetVariable("textord_tabfind_find_tables", "1");
        api.SetVariable("textord_tablefind_good_width", "3");
    }
    
    public static void configureForBusinessCard(TessBaseAPI api) {
        api.SetPageSegMode(PSM_SINGLE_BLOCK);
        api.SetVariable("preserve_interword_spaces", "1");
        api.SetVariable("tessedit_char_whitelist", 
                       "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789@.-+() ");
    }
    
    public static void configureForInvoice(TessBaseAPI api) {
        api.SetPageSegMode(PSM_AUTO);
        api.SetVariable("preserve_interword_spaces", "1");
        api.SetVariable("textord_tabfind_find_tables", "1");
    }
    
    public static void configureForLicensePlate(TessBaseAPI api) {
        api.SetPageSegMode(PSM_SINGLE_LINE);
        api.SetVariable("tessedit_char_whitelist", "ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789");
        api.SetVariable("classify_bln_numeric_mode", "0");
        api.SetVariable("textord_min_chars_in_word", "1");
    }
}

Types

Configuration Parameter Types

// Parameter value types
public class TessBaseAPI {
    // Integer parameters (thresholds, counts, modes)
    public boolean GetIntVariable(String name, int[] value);
    
    // Boolean parameters (feature toggles)
    public boolean GetBoolVariable(String name, boolean[] value);
    
    // Double parameters (floating-point thresholds)
    public boolean GetDoubleVariable(String name, double[] value);
    
    // String parameters (paths, character sets, languages)
    public String GetStringVariable(String name);
}

Engine Mode Constants

public static final int OEM_TESSERACT_ONLY = 0;           // Legacy engine
public static final int OEM_LSTM_ONLY = 1;                // LSTM neural network
public static final int OEM_TESSERACT_LSTM_COMBINED = 2;  // Combined (deprecated)
public static final int OEM_DEFAULT = 3;                  // Default mode

Page Segmentation Constants

public static final int PSM_OSD_ONLY = 0;
public static final int PSM_AUTO_OSD = 1;
public static final int PSM_AUTO_ONLY = 2;   
public static final int PSM_AUTO = 3;
public static final int PSM_SINGLE_COLUMN = 4;
public static final int PSM_SINGLE_BLOCK_VERT_TEXT = 5;
public static final int PSM_SINGLE_BLOCK = 6;
public static final int PSM_SINGLE_LINE = 7;
public static final int PSM_SINGLE_WORD = 8;
public static final int PSM_CIRCLE_WORD = 9;
public static final int PSM_SINGLE_CHAR = 10;
public static final int PSM_SPARSE_TEXT = 11;
public static final int PSM_SPARSE_TEXT_OSD = 12;
public static final int PSM_RAW_LINE = 13;

Install with Tessl CLI

npx tessl i tessl/maven-org-bytedeco--tesseract-platform

docs

configuration.md

core-ocr-engine.md

index.md

language-support.md

layout-analysis.md

output-renderers.md

result-navigation.md

tile.json