TextRecognitionResult<T>
Models & Types · AiDotNet.Document
Represents the result of text recognition from a cropped image.
For Beginners
Text recognition reads the actual characters from an image of text. This result contains the recognized text string along with confidence scores for each character.
Constructors
| Constructor | Summary |
|---|---|
TextRecognitionResult() |
Properties
| Property | Summary |
|---|---|
IReadOnlyList<CharacterRecognition<T>> Characters { get; set; } |
Gets the per-character confidence scores. |
IReadOnlyList<ValueTuple<string, double>> Alternatives { get; set; } |
Gets alternative recognition hypotheses with their scores. |
T Confidence { get; set; } |
Gets the overall confidence score (0-1). |
Tensor<T> AttentionWeights { get; set; } |
Gets the attention weights for visualization (if available). |
Tensor<T> CharacterProbabilities { get; set; } |
Gets the character-level probability distribution (shape: [seq_len, vocab_size]). |
double ConfidenceValue { get; set; } |
Gets the confidence as a double value. |
double ProcessingTimeMs { get; set; } |
Gets the processing time in milliseconds. |
string Text { get; set; } |
Gets the recognized text string. |