---
title: Monitait Complete User Guide
subtitle: WatcherJET 3.0 Hardware • AI Training Platform
author: Monitait.com
date: "2025-11-23"
format:
  html:
    grid:
      sidebar-width: 300px
      body-width: 900px
---

::: {.callout-note}
**Version:** 1.0 • **Date:** October 13, 2025  
**Powered by Monitait** – [monitait.com](https://monitait.com)
:::

## Download Full Illustrated PDFs

| Document                                    | Pages | Download |
|---------------------------------------------|-------|----------|
| Chapter 1 – Hardware & WatcherJET 3.0       | 27    | [CHAP-1-AI-HARDWARE.pdf](CHAP-1-AI-HARDWARE.pdf) |
| Chapter 2 – AI Training                     | 15    | [CHAP-2-AI-TRAINING.pdf](CHAP-2-AI-TRAINING.pdf) |

*All diagrams, wiring schematics, and screenshots are inside the PDFs.*

# Chapter 1 – Hardware & WatcherJET 3.0

### Quick Setup (4 Steps)
1. Connect the sensors  
2. Connect the power supply (12–24 V DC)  
3. Provide an access point  
4. Register at console.monitait.com/factory/watchers

### Step 1: Connect the Sensors

#### External Machine Signal
- **Production Count**: Any 12–24 V signal → OK inputs (3 & 4)  
- **Defect Count**: Ejector/NG signal → NG inputs (5 & 6)  
  → Bidirectional, opto-isolated

#### Push Button
Production: Button → OK(4) + GND(1), bridge OK(3) → +V(2)  
Defects: Same using NG(5 & 6)

#### Obstacle Sensor
Black → OK/NG input | Brown → +V(2) | Blue → GND(1) | Bridge remaining → +V(2)

#### Encoder
White → NG(6) | Black → OK(4) | Brown → +V | Blue → GND  
Bridge (3) & (5) → +V(2)

#### RS485
A → terminal 8 | B → terminal 7

### Step 2: Power Supply
12–24 V DC, max 2 A → terminals (1 & 2) → green power LED on

### Step 3: Network
**Best:** Wired LAN (allow `*.monitait.com`)  
**Temporary:** Hotspot → SSID: `Monitait`, Password: `p@ssword`

### Step 4: Register Watcher
1. console.monitait.com → Watchers → Add Watcher  
2. Enter Registration ID  
3. Test: trigger sensor → red heart LED blinks

### High-Current PSU, Emitters & Actuators
- High-current ≤48 V → terminals (9 & 10)  
- Emitters: Positive → PSU, Negative → U(12) or B(11)  
- Ejector (15), Warning (16) → PLC (opto-isolated NPN)

### Keys & Indicators
- Key-1: Buzzer on/off  
- Key-3: Camera enable  
- Key-4: Scanner enable  
- Green = power, Red heart = data, Checkmark = OK, Rejection = NG

# Chapter 2 – AI Training

### Step 1: AI Training Platform – Annotation
1. Confirm task  
2. Choose category  
3. Draw bounding box  
4. Save → next image

**Keyboard shortcuts** (very fast):  
`1 2 3 4 Q W E R T Y U I` → instant category selection

**Tips**
- Overlapping objects → draw elsewhere → drag  
- No images visible → all done, use < > buttons  
- Metadata → click “+” in right panel

### Step 2: Administration – Create Task
1. Tasks → **+ NEW TASK**  
   - Name, quantity, type, dates  
   - Define labels/categories with colors  
2. ⋮ → **Upload Images** (.jpg/.png/.jpeg)

**Best Practices**
- Visually distinct categories  
- Always label main object (e.g., “bottle”)  
- Use descriptive IDs (bottle-pk-100)  
- Monitait team handles training & augmentation

### Step 3: Evaluation

| Term  | Meaning                     | Formula                     |
|-------|-----------------------------|-----------------------------|
| TP    | Correct detection           | —                           |
| FP    | False alarm                 | —                           |
| FN    | Missed defect               | —                           |
| Precision | TP/(TP+FP)              | Accuracy of detections      |
| Recall    | TP/(TP+FN)              | % of real defects found     |
| F1-Score  | 2×(P×R)/(P+R)           | Balanced score              |
| mAP   | Mean Average Precision      | Overall quality (higher = better) |

**How to improve**
- Diverse lighting/angles/backgrounds  
- Tight, consistent bounding boxes  
- Multiple reviewers  
- Avoid similar categories  
- Let Monitait team choose best augmentations

### Step 4: Deploy Model
Successful training produces:
- `best.pt` (weights)
- Unique **Training ID** (record it!)

**On production machine:**
```bash
cp best.pt /home/projects/inference/best.pt
# restart inference service → new model live