About Prompt
- Prompt Type – Dynamic
- Prompt Platform – ChatGPT, Grok, Deepseek, Gemini, Copilot, Midjourney, Meta AI and more
- Niche – Visual Material Analysis
- Language – English
- Category – Material Quality Inspection
- Prompt Title – AI Prompt for Classifying Raw Material Quality Using Image Recognition
Prompt Details
This prompt is designed for classifying raw material quality using image recognition across various AI platforms specializing in visual material analysis. Its dynamic nature allows for customization based on specific material types, defects, and desired output.
**Prompt Structure:**
“`
## Task: Raw Material Quality Classification using Image Recognition
**1. Material Specification:**
* **Material Type:** [Insert specific material name, e.g., wood, steel, fabric, plastic, etc.]
* **Form:** [Specify the form of the material, e.g., planks, sheets, pellets, rolls, etc.]
* **Expected Quality Characteristics:** [List the key visual features that define good quality for this material, e.g., uniform color, smooth surface, no cracks, specific texture, etc.]
**2. Defect Specification (Optional, but highly recommended):**
* **Target Defects:** [List the specific defects to be identified, e.g., knots in wood, rust on steel, tears in fabric, discoloration in plastic, etc.]
* **Defect Severity Levels (Optional):** [Define levels of severity for each defect if applicable, e.g., minor, moderate, severe, critical, etc. Provide visual descriptions or examples for each level.]
**3. Image Input:**
* **Image:** [Provide the input image either as a file upload, URL, or base64 encoded string.]
* **Image Context (Optional but helpful):** [Provide any relevant information about the image acquisition, e.g., lighting conditions, camera angle, magnification, etc.]
**4. Output Format:**
Choose one or more of the following output formats:
* **Classification Label:** [Specify the desired output label. This could be a simple “Pass/Fail” or a more detailed quality grade, e.g., “Grade A”, “Grade B”, “Reject”.]
* **Defect Localization (if applicable):** [Request bounding boxes or segmentation masks around identified defects in the image.]
* **Confidence Score:** [Request a confidence score (e.g., 0-1 or 0-100%) for each classification or defect detection.]
* **Detailed Explanation (Optional):** [Request a textual explanation justifying the classification or detailing the identified defects and their severity.]
**5. Evaluation Metrics (Optional for model training and fine-tuning):**
* **Specify the desired evaluation metrics for the classification task.** [e.g., accuracy, precision, recall, F1-score, etc.]
**Example Prompt (Wood Inspection):**
## Task: Raw Material Quality Classification using Image Recognition
**1. Material Specification:**
* **Material Type:** Wood
* **Form:** Planks
* **Expected Quality Characteristics:** Uniform light brown color, smooth surface, no knots, straight grain.
**2. Defect Specification:**
* **Target Defects:** Knots, cracks, discoloration
* **Defect Severity Levels:**
* **Knots:** Minor (diameter < 5mm), Moderate (diameter 5-10mm), Severe (diameter > 10mm)
* **Cracks:** Minor (length < 2cm), Moderate (length 2-5cm), Severe (length > 5cm)
**3. Image Input:**
* **Image:** [Provide image URL or file upload]
* **Image Context:** Image taken under natural lighting conditions.
**4. Output Format:**
* **Classification Label:** Pass/Fail
* **Defect Localization:** Bounding boxes around identified defects.
* **Confidence Score:** For Pass/Fail classification and each detected defect.
* **Detailed Explanation:** List identified defects and their severity.
**Example Prompt (Steel Sheet Inspection):**
## Task: Raw Material Quality Classification using Image Recognition
**1. Material Specification:**
* **Material Type:** Steel
* **Form:** Sheet
* **Expected Quality Characteristics:** Uniform grey color, smooth surface, no rust, no dents.
**2. Defect Specification:**
* **Target Defects:** Rust, dents, scratches
**3. Image Input:**
* **Image:** [Provide image URL or file upload]
**4. Output Format:**
* **Classification Label:** Grade A / Grade B / Reject
* **Defect Localization:** Segmentation masks highlighting the defective areas.
* **Confidence Score:** For each classification label.
**Note:** This dynamic prompt template allows flexibility to adjust the specific material, defects, image input, and desired output format based on your individual needs and the capabilities of the chosen AI platform. Ensure to replace the bracketed placeholders with accurate and detailed information. For optimal performance, consider providing example images with labeled defects during model training and fine-tuning when possible.
“`