SiamCafe.net Blog
Technology

Midjourney Prompt Citizen Developer สร้าง Visual Content ด้วย AI

midjourney prompt citizen developer
Midjourney Prompt Citizen Developer | SiamCafe Blog
2025-07-06· อ. บอม — SiamCafe.net· 1,457 คำ

Midjourney ????????? Citizen Developer ?????????????????????

Midjourney ???????????? AI image generation tool ?????????????????????????????????????????? text prompts ???????????? Discord ????????? diffusion model ?????????????????????????????????????????????????????????????????? professional ???????????????????????????????????????????????????????????? ??????????????????????????????????????????????????????????????? generative AI ?????????????????????????????????????????????????????????????????????????????????????????? DALL-E ????????? Stable Diffusion

Citizen Developer ???????????????????????????????????????????????????????????????????????????????????????????????????????????????????????? applications ???????????? solutions ??????????????????????????????????????????????????? low-code/no-code ????????????????????????????????? AI image generation Citizen Developer ????????????????????????????????? marketing materials, product mockups, website designs, social media content ?????????????????????????????????????????? graphic designer

?????????????????? Midjourney ????????? Citizen Developer ????????????????????????????????????????????? marketing, product, sales ??????????????? visual content ?????????????????? ????????????????????????????????????????????????????????? ?????? rapid prototyping ???????????????????????? ??????????????? concept ???????????? invest ?????? production quality

????????????????????????????????? Prompt ?????????????????? Midjourney

????????????????????????????????? prompt ?????????????????????????????????????????????????????????

# === Midjourney Prompt Engineering ===

cat > prompt_guide.yaml << 'EOF'
prompt_structure:
  format: "[Subject] [Style] [Details] [Parameters]"
  
  components:
    subject:
      description: "??????????????????????????????????????????????????????????????????????????????"
      examples:
        - "a beautiful Thai temple at sunset"
        - "a futuristic city with flying cars"
        - "a cozy coffee shop interior"
      tips:
        - "????????????????????????????????? ???????????????????????????????????????????????????"
        - "????????? adjectives ????????????????????????????????????"
        
    style:
      description: "?????????????????????????????????"
      options:
        - "photorealistic, 8k, ultra detailed"
        - "watercolor painting, soft colors"
        - "3D render, Pixar style"
        - "minimalist flat design, vector"
        - "oil painting, impressionist"
        - "anime style, Studio Ghibli"
        - "cyberpunk, neon lights"
        - "vintage photograph, 1970s"
        
    lighting:
      options:
        - "golden hour lighting"
        - "dramatic lighting, chiaroscuro"
        - "soft studio lighting"
        - "neon glow, backlit"
        - "natural daylight"
        
    camera:
      options:
        - "wide angle lens"
        - "macro photography"
        - "aerial view, drone shot"
        - "portrait, shallow depth of field"
        - "fisheye lens"
        
    parameters:
      ar: "Aspect ratio (--ar 16:9, --ar 1:1, --ar 9:16)"
      v: "Version (--v 6.1)"
      q: "Quality (--q 2 for highest)"
      s: "Stylize (--s 100 default, --s 750 artistic)"
      c: "Chaos (--c 0-100, higher = more varied)"
      no: "Negative prompt (--no text, logo)"
      
  example_prompts:
    marketing:
      - "modern minimalist product photography, white skincare bottle, marble surface, soft studio lighting, clean background --ar 1:1 --v 6.1 --s 200"
    social_media:
      - "flat lay photography, Thai street food, top down view, warm colors, food styling --ar 4:5 --v 6.1"
    website_hero:
      - "abstract gradient background, purple and blue, flowing shapes, modern, clean --ar 16:9 --v 6.1 --s 400"
EOF

echo "Prompt guide created"

??????????????? Prompt Library ???????????? Python

?????????????????????????????? prompt templates

#!/usr/bin/env python3
# prompt_library.py ??? Midjourney Prompt Library
import json
import logging
import random
from typing import Dict, List

logging.basicConfig(level=logging.INFO)
logger = logging.getLogger("prompts")

class PromptLibrary:
    def __init__(self):
        self.templates = {}
        self.history = []
    
    def add_template(self, category, name, template, params=None):
        if category not in self.templates:
            self.templates[category] = {}
        self.templates[category][name] = {
            "template": template,
            "params": params or {"ar": "1:1", "v": "6.1", "s": "200"},
        }
    
    def generate_prompt(self, category, name, variables=None):
        """Generate prompt from template"""
        if category not in self.templates or name not in self.templates[category]:
            return {"error": "Template not found"}
        
        tmpl = self.templates[category][name]
        prompt = tmpl["template"]
        
        if variables:
            for key, value in variables.items():
                prompt = prompt.replace(f"{{{key}}}", value)
        
        params = " ".join(f"--{k} {v}" for k, v in tmpl["params"].items())
        full_prompt = f"{prompt} {params}"
        
        self.history.append({"category": category, "name": name, "prompt": full_prompt})
        return {"prompt": full_prompt}
    
    def batch_generate(self, category, name, variations):
        """Generate multiple variations"""
        results = []
        for var in variations:
            result = self.generate_prompt(category, name, var)
            if "prompt" in result:
                results.append(result["prompt"])
        return results
    
    def seed_library(self):
        """Pre-populate with useful templates"""
        self.add_template("marketing", "product_shot",
            "{product} on {surface}, professional product photography, studio lighting, clean background",
            {"ar": "1:1", "v": "6.1", "s": "200", "q": "2"})
        
        self.add_template("marketing", "hero_banner",
            "{scene}, cinematic composition, {mood} atmosphere, wide angle",
            {"ar": "16:9", "v": "6.1", "s": "300"})
        
        self.add_template("social", "instagram_post",
            "{subject}, aesthetic photography, {style}, trending on instagram",
            {"ar": "4:5", "v": "6.1", "s": "250"})
        
        self.add_template("ui_design", "app_mockup",
            "modern mobile app UI design, {screen_type}, minimalist, {color_scheme} color palette, Figma style",
            {"ar": "9:16", "v": "6.1", "s": "150"})
        
        self.add_template("presentation", "slide_bg",
            "abstract background, {style}, {colors}, subtle texture, professional",
            {"ar": "16:9", "v": "6.1", "s": "400", "no": "text"})

lib = PromptLibrary()
lib.seed_library()

# Generate prompts
product = lib.generate_prompt("marketing", "product_shot", {
    "product": "white ceramic coffee mug",
    "surface": "wooden table with plants",
})
print(f"Product: {product['prompt']}")

hero = lib.generate_prompt("marketing", "hero_banner", {
    "scene": "Thai temple at golden hour",
    "mood": "peaceful and serene",
})
print(f"Hero: {hero['prompt']}")

# Batch variations
variations = lib.batch_generate("social", "instagram_post", [
    {"subject": "Thai street food pad thai", "style": "warm tones, food photography"},
    {"subject": "Bangkok skyline at night", "style": "cyberpunk, neon lights"},
    {"subject": "tropical beach Phuket", "style": "travel photography, vivid colors"},
])
print(f"\nBatch: {len(variations)} prompts generated")

Citizen Developer ????????? AI Image Generation

Use cases ?????????????????? citizen developers

#!/usr/bin/env python3
# citizen_dev_workflow.py ??? Citizen Developer AI Workflow
import json
import logging
from typing import Dict, List

logging.basicConfig(level=logging.INFO)
logger = logging.getLogger("workflow")

class CitizenDevWorkflow:
    def __init__(self):
        self.workflows = {}
    
    def use_cases(self):
        return {
            "marketing_team": {
                "tasks": [
                    "??????????????? social media graphics (IG, FB, TikTok)",
                    "Product mockup ?????????????????? pitch deck",
                    "Email newsletter header images",
                    "Blog post featured images",
                    "Ad creative variations ?????????????????? A/B testing",
                ],
                "tools": ["Midjourney", "Canva", "Figma"],
                "time_saved": "70-80% ???????????????????????????????????? designer",
                "cost_saved": "5,000-50,000 THB/???????????????",
            },
            "product_team": {
                "tasks": [
                    "Rapid prototyping UI concepts",
                    "User persona illustrations",
                    "Product feature visualizations",
                    "Presentation slides backgrounds",
                    "Wireframe to mockup conversion",
                ],
                "tools": ["Midjourney", "Figma", "Notion AI"],
                "time_saved": "50-60%",
            },
            "content_team": {
                "tasks": [
                    "Blog illustrations",
                    "Infographic elements",
                    "E-book covers",
                    "YouTube thumbnails",
                    "Podcast cover art",
                ],
                "tools": ["Midjourney", "DALL-E", "Canva"],
                "time_saved": "60-70%",
            },
            "hr_team": {
                "tasks": [
                    "Job posting graphics",
                    "Company culture illustrations",
                    "Training material visuals",
                    "Employee handbook graphics",
                ],
                "tools": ["Midjourney", "Canva"],
                "time_saved": "40-50%",
            },
        }
    
    def no_code_integration(self):
        return {
            "zapier": {
                "workflow": "Google Form ??? Midjourney ??? Google Drive ??? Slack notification",
                "use_case": "????????? marketing ???????????? form ??????????????? ??????????????????????????????????????????????????????",
            },
            "make_com": {
                "workflow": "Airtable ??? Generate prompt ??? Midjourney API ??? Upload to CMS",
                "use_case": "Batch generate blog images ????????? content calendar",
            },
            "n8n": {
                "workflow": "Webhook ??? AI prompt ??? Image generate ??? Resize ??? Upload",
                "use_case": "Self-hosted automation pipeline",
            },
        }

workflow = CitizenDevWorkflow()
cases = workflow.use_cases()
print("Citizen Developer Use Cases:")
for team, info in cases.items():
    print(f"\n  {team}:")
    print(f"    Tasks: {len(info['tasks'])}")
    print(f"    Time saved: {info['time_saved']}")
    for task in info["tasks"][:2]:
        print(f"    - {task}")

Workflow ????????????????????????????????????????????? Image Generation

??????????????? automated image generation pipeline

# === Automated Image Generation Workflow ===

# 1. Content Calendar ??? Auto Generate Images
cat > auto_generate.py << 'PYEOF'
#!/usr/bin/env python3
"""Auto-generate images from content calendar"""
import json
import csv
from datetime import datetime

class ImageAutoGenerator:
    def __init__(self):
        self.queue = []
    
    def load_calendar(self, calendar_data):
        """Load content calendar and create image requests"""
        for entry in calendar_data:
            prompt = self.create_prompt(entry)
            self.queue.append({
                "title": entry["title"],
                "prompt": prompt,
                "size": entry.get("size", "1:1"),
                "due_date": entry["publish_date"],
                "status": "pending",
            })
        return f"Loaded {len(self.queue)} image requests"
    
    def create_prompt(self, entry):
        """Create Midjourney prompt from content metadata"""
        category = entry.get("category", "general")
        
        style_map = {
            "tech": "modern, futuristic, clean design, blue tones",
            "food": "food photography, warm lighting, appetizing, top view",
            "travel": "travel photography, vivid colors, wide angle, golden hour",
            "business": "corporate, professional, modern office, clean",
            "lifestyle": "lifestyle photography, natural light, aesthetic",
        }
        
        style = style_map.get(category, "professional photography")
        return f"{entry['title']}, {style}, high quality --ar {entry.get('size', '16:9')} --v 6.1"
    
    def process_queue(self):
        """Process image generation queue"""
        results = []
        for item in self.queue:
            # In production: call Midjourney API
            item["status"] = "generated"
            results.append(item)
        return results

# Example usage
calendar = [
    {"title": "Thai street food guide", "category": "food", "publish_date": "2024-07-01", "size": "16:9"},
    {"title": "Best laptops 2024", "category": "tech", "publish_date": "2024-07-03", "size": "16:9"},
    {"title": "Phuket travel tips", "category": "travel", "publish_date": "2024-07-05", "size": "4:5"},
]

gen = ImageAutoGenerator()
print(gen.load_calendar(calendar))

for item in gen.queue:
    print(f"  {item['title']}: {item['prompt'][:60]}...")
PYEOF

# 2. Batch Processing Script
cat > batch_images.sh << 'BASH'
#!/bin/bash
# Batch image generation and processing

INPUT_DIR="./prompts"
OUTPUT_DIR="./generated"
PROCESSED_DIR="./processed"

mkdir -p $OUTPUT_DIR $PROCESSED_DIR

echo "=== Batch Image Processing ==="

# Resize for different platforms
for img in $OUTPUT_DIR/*.png; do
    filename=$(basename "$img" .png)
    
    # Instagram (1080x1080)
    convert "$img" -resize 1080x1080^ -gravity center -extent 1080x1080 \
        "$PROCESSED_DIR/_ig.jpg"
    
    # Facebook (1200x630)
    convert "$img" -resize 1200x630^ -gravity center -extent 1200x630 \
        "$PROCESSED_DIR/_fb.jpg"
    
    # Blog (1200x600)
    convert "$img" -resize 1200x600^ -gravity center -extent 1200x600 \
        "$PROCESSED_DIR/_blog.jpg"
    
    echo "Processed: $filename"
done

echo "Done! $(ls $PROCESSED_DIR | wc -l) images processed"
BASH
chmod +x batch_images.sh

echo "Automation workflow created"

Best Practices ????????? Copyright

??????????????????????????????????????????????????????

#!/usr/bin/env python3
# best_practices.py ??? AI Image Generation Best Practices
import json
import logging

logging.basicConfig(level=logging.INFO)
logger = logging.getLogger("practices")

class BestPractices:
    def __init__(self):
        pass
    
    def copyright_guidelines(self):
        return {
            "midjourney_terms": {
                "paid_plan": "??????????????????????????????????????????????????????????????????????????? ???????????????????????????????????????????????????",
                "free_plan": "Midjourney ?????? CC BY-NC 4.0 license",
                "important": "???????????????????????????????????????????????????????????????????????????????????????????????????????????????",
            },
            "do": [
                "?????????????????? AI ?????????????????? concept, mockup, social media",
                "???????????? prompt ??????????????????????????????????????????????????? reference",
                "?????????????????????????????? AI ????????????????????? (??????????????? artifacts)",
                "?????????????????????????????????????????? AI generated (???????????????????????????)",
                "????????? paid plan ??????????????????????????? commercial",
            ],
            "dont": [
                "?????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????",
                "??????????????????????????? deepfake ????????????????????????????????????????????????",
                "??????????????????????????? content ????????????????????????????????????",
                "?????????????????????????????? AI ?????????????????? photography ???????????? 100%",
                "?????????????????????????????????????????? AI ?????????????????????????????????????????????",
            ],
        }
    
    def quality_checklist(self):
        return {
            "before_publishing": [
                "?????????????????? AI artifacts (?????????????????????????????????????????? ??????????????????????????????????????????)",
                "????????????????????? brand consistency (?????? ????????? ???????????????)",
                "Resize ????????????????????????????????? platform",
                "Compress ?????????????????? web (WebP/AVIF)",
                "??????????????? alt text ?????????????????? accessibility",
                "????????????????????? copyright compliance",
            ],
        }

practices = BestPractices()
guidelines = practices.copyright_guidelines()
print("Copyright Guidelines:")
print(f"  Paid plan: {guidelines['midjourney_terms']['paid_plan']}")
print("\nDo:")
for item in guidelines["do"][:3]:
    print(f"  + {item}")
print("\nDon't:")
for item in guidelines["dont"][:3]:
    print(f"  - {item}")

FAQ ??????????????????????????????????????????

Q: Midjourney ????????? DALL-E ????????? Stable Diffusion ???????????????????????????????????????????

A: ???????????????????????????????????????????????????????????????????????? Midjourney ???????????????????????????????????? ??????????????????????????? default ????????????????????????????????? artistic, marketing, creative work ????????????????????? Discord ???????????? $10-60/??????????????? DALL-E 3 (ChatGPT) ?????????????????????????????? ???????????????????????????????????????????????????????????? ?????? safety filters ????????????????????? ????????????????????? ChatGPT Plus $20/??????????????? Stable Diffusion open source ????????? customize ???????????????????????????????????? ?????????????????? GPU ????????????????????? cloud ??????????????? developer ?????????????????? Citizen Developer ??????????????? Midjourney (??????????????????) ???????????? DALL-E (????????????) ?????????????????? developer ??????????????? Stable Diffusion (???????????????????????? ?????????)

Q: Citizen Developer ??????????????????????????????????????????????????????????

A: ?????????????????????????????????????????? Prompt Engineering ??????????????? prompt ????????????????????????????????????????????????????????????????????? ??????????????????????????????????????? 1-2 ?????????????????????, Basic Design ?????????????????? composition ?????? layout ????????????????????????????????? Canva templates, No-Code Tools ????????? Zapier, Make.com ??????????????? automation, Project Management ?????????????????? content calendar ??????????????? deadline ??????????????????????????? coding, graphic design ?????????????????????, AI/ML knowledge ????????????????????????????????? YouTube, online courses ?????????????????? ???????????????????????? Midjourney + Canva ???????????? combo ??????????????????????????????????????????????????? beginners

Q: ?????????????????? AI ?????????????????? commercial ???????????????????

A: ????????? ?????????????????? paid plan Midjourney paid plan ($10+/???????????????) ?????????????????????????????????????????????????????????????????????????????? DALL-E (ChatGPT Plus) ???????????????????????????????????????????????????????????? Stable Diffusion open source ????????????????????????????????? ????????????????????????????????? ???????????????????????????????????? > $1M/?????? ????????????????????? Midjourney Pro ($60/???????????????) ????????????????????????????????????????????????????????????????????? copyright ??????????????????????????? ??????????????????????????????????????????????????????????????????????????? (advertising, publishing) ????????????????????? terms of service ???????????????????????? platform ????????????

Q: ??????????????????????????? prompt ?????????????????????????????????????

A: ????????????????????? ????????????????????????????????? ????????????????????????????????????????????? "a cat" ??????????????????????????? "a fluffy orange tabby cat sitting on a windowsill", ??????????????? style descriptor ???????????? "photorealistic, 8k, cinematic lighting", ????????? reference ?????????????????????????????? photographer (???????????? "in the style of Studio Ghibli"), ???????????? composition (wide angle, close-up, aerial view), ????????? negative prompt --no ????????????????????????????????????????????????????????????, ??????????????? parameters --s (stylize), --c (chaos), --ar (aspect ratio), ???????????? prompt ?????????????????????????????? reuse ?????????

📖 บทความที่เกี่ยวข้อง

Midjourney Prompt Microservices Architectureอ่านบทความ → Midjourney Prompt Service Mesh Setupอ่านบทความ → Elixir Ecto Citizen Developerอ่านบทความ → Midjourney Prompt อ่านบทความ → Midjourney Prompt Hexagonal Architectureอ่านบทความ →

📚 ดูบทความทั้งหมด →