Update README.md

README.md

@@ -1,48 +1,187 @@
 ---
-title: Docs
-emoji: 🌖
-colorFrom: green
-colorTo: pink
-sdk: docker
-pinned: false
+title: "Monitait Step-by-Step User Guide"
+subtitle: "Hardware (WatcherJET 3.0) & AI Training Platform"
+author: "Monitait.com"
+date: "2025-10-13"
+format:
+  html:
+    toc: true
+    toc-depth: 4
+    toc-location: left
+    theme: cosmo
+    code-fold: true
+    grid:
+      sidebar-width: 300px
+      body-width: 900px
+      margin-width: 300px
+  pdf:
+    toc: true
+    keep-tex: true
+    documentclass: scrreprt
+    classoption: ["paper=a4", "twoside=false"]
 ---
 
-To get started working with quarto we recommend you first [install quarto](https://quarto.org/docs/get-started/) locally so that you can render the site without Docker.
-We also recommend the [Quarto VS Code Extension](https://marketplace.visualstudio.com/items?itemName=quarto.quarto) which provides syntax highlighting, code completion, a preview button and more.
+::: {.callout-note}
+**Version:** 1.0 • **Date:** October 13, 2025  
+**Powered by Monitait** – monitait.com
+:::
 
-The quarto source is located in `src` and you can preview the site with:
+# Chapter 1 – Hardware & WatcherJET 3.0
 
-```
-quarto preview src
-```
+![WatcherJET 3.0 Overview](images/watcherjet-cover.png){width=80% fig-align=center}
 
-A web browser should open up with a live preview of the site.
+## Table of Contents – Setup: Collecting Data
 
-## Making changes
+- Step 1: Connect the sensors
+  - External machine signal
+  - Push button
+  - Obstacle sensor
+  - Encoder
+  - RS485 protocol
+- Step 2: Connect the power supply
+- Step 3: Provide an access point
+- Step 4: Register at console.monitait.com
 
-The `src/_quarto.yml` contains the site-level configuration for the quarto website and tells quarto which files to render, and how they should be organized.
-For example if you wanted to modify the [site navigation](https://quarto.org/docs/reference/site-navigation.html) you should modify this file.
+## Table of Contents – Setup: Taking Action
 
-Quarto can render markdown, ipynb, and .qmd files, and you can mix formats in a single document.
+- Step 1: Connect the high-current power supply
+- Step 2: Set up the emitters
+- Step 3: Connect the actuators
+- Controls and Signals (Keys & Indicators)
+- QC Machines & Wiring Schematics
 
-## Executing code
+### Step 1: Connect the Sensors
 
-One of the main virtues of Quarto is that it lets you combine code and text in a single document.
-By default if you include a code chunk in your document, Quarto will execute that code and include the output in the rendered document.
-This is great for reproducibility and for creating documents that are always up-to-date.
+#### External Machine Signal
+![External Machine Signal](images/sensors-external.png){width=90%}
 
-```{python}
-import seaborn as sns
-import matplotlib.pyplot as plt
+- **Production Count**: Connect any 12–24 V signal to OK inputs (3 & 4)  
+- **Counting Defects**: Connect ejector/machine NG signal to NG inputs (5 & 6)  
+  → Bidirectional & opto-isolated
 
-# Sample data
-tips = sns.load_dataset("tips")
+#### Push Button
+![Push Button Wiring](images/sensors-pushbutton.png){width=90%}
 
-# Create a seaborn plot
-sns.set_style("whitegrid")
-g = sns.lmplot(x="total_bill", y="tip", data=tips, aspect=2)
-g = (g.set_axis_labels("Total bill (USD)", "Tip").set(xlim=(0, 60), ylim=(0, 12)))
+**Production Count** → OK (4) + GND (1), bridge OK (3) → +V (2)  
+**Defect Count** → same logic using NG (6 & 5)
 
-plt.title("Tip by Total Bill")
-plt.show()
-```
+#### Obstacle Sensor
+![Obstacle Sensor Wiring](images/sensors-obstacle.png){width=90%}
+
+Black → OK/NG input, Brown → +V (2), Blue → GND (1), bridge remaining OK/NG → +V (2)
+
+#### Encoder & RS485
+Encoder: White → NG (6), Black → OK (4), Brown → +V, Blue → GND  
+RS485: A → (8), B → (7)
+
+### Step 2: Connect the Power Supply
+**Specifications**
+
+| Parameter              | Value           |
+|------------------------|-----------------|
+| Input/Output Voltage   | 12–24 V DC      |
+| Max Output Current     | 2 A             |
+| Operating Temperature  | –10 °C to 50 °C |
+
+### Step 3: Provide an Access Point
+**Best Practice:** Wired LAN + firewall rule `*.monitait.com`  
+**Temporary:** Mobile hotspot → Name: `Monitait`, Password: `p@ssword`
+
+### Step 4: Register at console.monitait.com
+Add Watcher → Enter Registration ID → Set station & advanced options (multiplication factor, timeout, etc.)
+
+# Chapter 2 – AI Training
+
+![AI Training Cover](images/ai-training-cover.png){width=70% fig-align=center}
+
+## Table of Contents
+
+- Step 1: AI Training Platform
+- Step 2: Administration (Create tasks & upload images)
+- Step 3: Evaluation
+- Step 4: Deploy trained weights
+
+### Step 1: AI Training Platform
+
+![Task List](images/ai-task-list.png){width=95%}
+
+#### Basic Tools – How to Annotate
+![Basic Annotation](images/ai-basic-tools.png){width=95%}
+
+1. Confirm correct task  
+2. Choose category  
+3. Draw bounding box  
+4. Save → next image
+
+#### Advanced Tools
+![Advanced Tools](images/ai-advanced-tools.png){width=95%}
+
+- Image ID display
+- Show/Hide panels
+- Pan & zoom (Ctrl + scroll)
+- Edit / delete boxes
+
+#### Tips & Hints
+![Tips](images/ai-tips.png){width=90%}
+
+- Keyboard shortcuts: 1–9, QWERTYUI for categories
+- Overlapping objects → draw elsewhere → drag, or hide with eye icon
+- Add metadata via “+” in right panel
+
+### Step 2: Administration – Create Tasks & Upload Images
+
+1. Tasks → **+ NEW TASK**
+   - Basic Info (name, quantity, type, dates, etc.)
+   - Labels → define categories & colors
+2. After creation → ⋮ → **Upload Images** (.jpg, .png, .jpeg)
+
+**Best Practices**
+- Use visually distinct categories
+- Always label main object (e.g., “bottle”)
+- Use descriptive IDs (bottle-pk-100)
+- Monitait team handles training & augmentation
+
+### Step 3: Evaluation
+
+#### What is Evaluation?
+Testing the model on unseen data → measures real-world performance.
+
+#### Key Metrics
+- **TP** – correctly detected defect
+- **FP** – false alarm
+- **FN** – missed defect
+- **TN** – correctly identified good item
+
+| Metric   | Formula                          | Meaning                                  |
+|----------|----------------------------------|------------------------------------------|
+| Precision| TP / (TP + FP)                   | How many detections were correct         |
+| Recall   | TP / (TP + FN)                   | How many real defects were found         |
+| F1-Score | 2 × (P × R) / (P + R)            | Balanced score                           |
+| mAP      | Mean Average Precision (0–1)     | Overall model quality (higher = better)  |
+
+#### Tips to Increase Accuracy & Recall
+- Diverse dataset (lighting, angles, backgrounds)
+- Tight, consistent bounding boxes
+- Multiple reviewers
+- Avoid similar/confusing categories
+- Consult Monitait team for optimal augmentations
+
+### Step 4: Deploy Trained Model
+
+Each successful training creates:
+- `best.pt` (final weights)
+- Unique **Training ID** (record it!)
+
+**Deployment Steps**
+1. Verify new `best.pt` performs better on sample images
+2. Copy to machine:  
+   `/home/projects/inference/best.pt`
+3. Restart the inference system
+
+The new model is now live.
+
+---
+
+::: {.callout-tip}
+For technical support or custom training strategies, contact the Monitait team at **monitait.com**
+:::