ANY DOCUMENT SPLIT API

• splits — a list of the detected document segments. Each segment is an array of two numbers, where the first number is the start page of the segment and the second number is the end page of the segment (1-based and inclusive). The segments are returned in the original page order and together cover every page of the upload.
• meta — additional information about the split. It currently contains emptyPages, a list of the 1-based page indexes that were detected as empty/blank (for example the blank backs of single-sided sheets scanned in duplex). These pages are folded into the preceding document and were skipped when deciding where to split. The list is empty when no empty pages were detected.

Suppose we have a document with 5 pages which needs to be splitted. We can split it into two segments: the first segment contains the first 3 pages and the second segment contains the last 2 pages. In this example page 3 was detected as an empty page, so it is reported under meta.emptyPages while still being part of the first segment.

An example result can look like this:

                                
                                {
                                    "splits": [
                                        [
                                            1,
                                            3
                                        ],
                                        [
                                            4,
                                            5
                                        ]
                                    ],
                                    "meta": {
                                        "emptyPages": [
                                            3
                                        ]
                                    }
                                }