> For the complete documentation index, see [llms.txt](https://docs.didge.io/home/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.didge.io/home/thermal-imagery-ai/admin/core-functions/didge-set-up/form-set-up.md).

# Form Set Up

Setting Up an AI-Enabled Image Data Extraction Form in the Didge Platform

Applies To: Didge Platform (Form Builder)

Purpose: To configure a form that enables image upload and AI-based data extraction into designated fields using ChatGPT prompts.

***

### 1. Purpose

This SOP outlines the steps to build a form in the Didge Platform that accepts an image input and uses ChatGPT to extract relevant information, populate designated fields, and store the output for operational workflows.

***

### 2. Prerequisites

* Access to the Didge Platform (form builder module)
* A valid Didge user account with form creation permissions
* An understanding of the information to be extracted from the image (e.g. object attributes, text data, identifiers, etc.)

***

### 3. Form Creation Procedure

#### Step 1: Start a New Form

* Log in to the Didge Platform.
* Navigate to Forms > Create New Form.
* Enter a form title and description relevant to the context (e.g., “Apple Type Identifier” or “Thermal Image Inspection”).<br>

#### Step 2: Insert a File Upload Field

* Drag and drop the “File Upload” field into the form.
* Label it clearly (e.g., “Image Upload”).
* Ensure file types accepted include JPG, PNG, and PDF (if scanning documents).
* Set the field to Required if every submission needs an image.

> Note: This image is the source for AI extraction. It can be uploaded manually or automatically by external integrations (e.g., mobile device + IFTTT).<br>

#### Step 3: Add Extraction Fields (Text/Number)

* Drag and drop Text Field, Number Field, or Date Field components as needed for the information you wish to extract.
* Label each field appropriately (e.g., “Apple Type”, “Surface Temperature”, “Document Date”).<br>

#### Step 4: Configure AI Prompt via Field Description

* For each field, open Field Settings > Description.
* Enter a clear, descriptive prompt that tells ChatGPT what to extract from the image.<br>

Examples:

<figure><img src="https://2326390636-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FhlcpP8RGwSYvfz0P333y%2Fuploads%2FnXyACHSm7l1PcL6LohBR%2FScreenshot%202025-05-12%20at%2010.02.30%E2%80%AFpm.png?alt=media&amp;token=a197a933-9e9d-4e0f-a1d0-b95f36fab0f4" alt=""><figcaption></figcaption></figure>

| Field Label     | Description (Prompt for AI)                         |
| --------------- | --------------------------------------------------- |
| Apple Type      | What is the apple type in the image?                |
| Surface Temp    | What is the surface temperature shown in the image? |
| Document Number | What is the invoice number on the scanned document? |

> Tip: Think of the description as a direct query to the AI.

#### Step 5: Save the Form

* Review all fields for correct labels, descriptions, and formats.
* Click Save Form.

***

### 4. Integration into Operations

#### Step 6: Link Form to Operation

* Navigate to Operations > Add Operation.
* Assign the form to a workflow or inspection where AI image extraction is needed.
* Optionally configure the operation to auto-launch from mobile or trigger via integration (e.g., thermal camera upload).

<figure><img src="https://2326390636-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FhlcpP8RGwSYvfz0P333y%2Fuploads%2F4OahXdmJKJCIT9UeksKo%2FScreenshot%202025-05-12%20at%2010.02.46%E2%80%AFpm.png?alt=media&amp;token=5a0316e1-aa8a-47a9-a449-9e09aebcaa82" alt=""><figcaption></figcaption></figure>

***

### 5. Use Case Notes & Best Practices

* Each field can extract distinct pieces of data from the same image.
* Add as many fields as required — the more prompts, the richer the extracted data.
* Ensure descriptions are context-specific and precise for optimal AI performance.
* Ideal for:
  * Product inspections (e.g., food, parts, documents)
  * OCR from technical manuals
  * Thermal image interpretations
  * Form digitization from photos

***

### 6. Verification and Testing

* Test the form by uploading an image manually.
* Observe how ChatGPT populates each field based on the prompt.
* Check that the data returned is:
  * Contextually accurate
  * Correctly formatted
  * Automatically submitted to the linked Didge database

***

### 7. Maintenance & Updates

* Update field descriptions as extraction requirements evolve.
* Periodically review AI performance on sample images.
* Maintain version control if forms are revised over time.

***


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.didge.io/home/thermal-imagery-ai/admin/core-functions/didge-set-up/form-set-up.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.