SOL & AI
In general documents and photos are stored as simple files on repositories, now SOL will read each file and extract all the possible information depending on the document category.
It can extract main metadata depending on the category of the document and any other information that is critical for the customer.
It also can extract the text from the labels visible on a equipment photo to recognize the model and serial number.
All this information is then associated to each document and searchable by a powerful searching engine allowing every user to easily find the information or document he is looking for.
Documents
- Pre-process documents with image quality detection
- Extract text and layout information from document files
- Identify key-value pairs in documents
- Split and classify documents by type
- Extract and normalize entities
- Label and review documents
- Obtain summaries and request internal information
- Store, search, organize, manage, and analyze documents and metadata
Images
- General description of the image, in the local language
- Classification of images with specific TAGs
- Extraction of all included text (equipment brands, serial numbers, barcodes, generic text)
- Definition of specific schemes for identifying faults or incorrect installations Automatic identification of objects, faces, text
Videos
- General description of the video, in the local language
- Transcription of verbal comments present in the video
- Classification of the video with TAGs
- Definition of specific schemes for identifying faults or incorrect installations
- Automatic identification of objects, faces, text, or voice
- Speech to text
