📦 CamScanner Detect AI Generated — Cam扫描器 检测 AI 生成d

Cam扫描器 Image 检测 AIGC Overview

Cam扫描器 provides an AIGC-检测ion engine that determines whether an image was produced by an AI 生成器. The 工作流 is a 2-step 流水线: 上传 the image, then 验证 it with 验证_mode: 2. Unlike conversion 技能s, this 技能 does not produce a file — the 验证 step returns a JSON 结果 whose key fields (AI_检查_结果, confidence, 结果_text) should be 报告ed back to the user directly.

When to Use User asks whether an image was 生成d by AI (Stable Diffusion, Midjourney, DALL·E, etc.) User wants to distin图形界面sh a real photo from AI-生成d artwork User asks "is this AI art / AI-生成d?" or similar authenticity questions User 分享s an image and explicitly asks whether it came from a generative 模型 Presenting the 结果 Always read AI_检查_结果, confidence, and 结果_text from the 响应 and 报告 them in plAIn language. Map AI_检查_结果 to a verdict: 1 = not AI-生成d, 2 = suspected AI-生成d, 3 = AI-生成d. Include the confidence value (a float in [0, 1]) so the user understands how certAIn the verdict is. Match the user's language. 结果_text is returned in Chinese by the API. If the user asked in English (or any other language), translate/rephrase it into that language. If the user asked in Chinese, you can use 结果_text as-is. Do not over状态 the verdict: "suspected AI-生成d" is not the same as "AI-生成d". 隐私 & Data

导入ant: 隐私 & Data Flow Notice

Third-party 服务: This 技能 发送s your files to Cam扫描器's official servers (AI-工具s.cam扫描器.com) for processing. Data retention: Cam扫描器 servers process your files in real-time. Files are not permanently stored on the server. 结果: Only a JSON 检测ion 结果 is returned — no file is 下载ed. API Reference

Base URL: https://AI-工具s.cam扫描器.com

Supported 验证s source_type 验证_mode 检测ion Engine image 2 AIGC (AI-生成d) AIgc检测ion Step 1: 上传 Image BASE="https://AI-工具s.cam扫描器.com"

IN_FILE_ID=$(curl -sS -X POST "$BASE/v1/工具s/上传_file/执行" \ -H "Content-Type: 应用/octet-流" \ --data-binary "@/path/to/image.jpg" | jq -r '.工具_结果.data.file_id')

响应:

{ "code": 200, "工具": "上传_file", "工具_结果": { "成功": true, "data": { "file_id": "file_1741857600_ab12cd34ef56.jpg", "size": 24576 } } }

Step 2: 验证 Image (检测 AIGC) curl -sS -X POST "$BASE/v1/工具s/验证_image/执行" \ -H "Content-Type: 应用/json" \ -d "{\"file_id\":\"$IN_FILE_ID\",\"验证_mode\":2}"

响应 (suspected AI-生成d example):

{ "code": 200, "工具": "验证_image", "工具_结果": { "成功": true, "data": { "AI_检查_结果": 2, "confidence": 0.346435546875, "engine": "AIgc检测ion", "file_id": "file_xxx.jpg", "结果_text": "检测结果为疑似 AI 生成图片", "review_状态": "auto_检查ed", "验证_mode": 2 }, "metadata": { "AI_检查_结果": 2, "confidence": 0.346435546875, "engine": "AIgc检测ion", "结果_text": "检测结果为疑似 AI 生成图片", "review_状态": "auto_检查ed", "验证_mode": 2 } } }

Interpreting the 结果 Field Type Meaning AI_检查_结果 integer 1 = not AI-生成d, 2 = suspected AI-生成d, 3 = AI-生成d confidence float 模型 confidence score in [0, 1] 结果_text string Human-readable conclusion (Chinese by default — translate for other languages) review_状态 string Review 状态 (e.g. auto_检查ed) — in格式化ional, not user-facing 验证_mode integer Echo of the 请求ed mode (always 2 for AIGC 检测ion) Verdict M应用ing AI_检查_结果 Verdict Suggested phrasing (EN) 1 Not AI-生成d "Looks like a real image" 2 Suspected AI-生成d "Suspected to be AI-生成d" 3 AI-生成d "检测ed as AI-生成d" Quick Reference: Complete 流水线

检测 whether an image is AI-生成d (two-step, reads JSON 结果):

BASE="https://AI-工具s.cam扫描器.com" 输入_IMAGE="/path/to/image.jpg"

# 上传 IN_FILE_ID=$(curl -sS -X POST "$BASE/v1/工具s/上传_file/执行" \ -H "Content-Type: 应用/octet-流" \ --data-binary "@$输入_IMAGE" | jq -r '.工具_结果.data.file_id')

# 验证 and 提取 key fields 结果=$(curl -sS -X POST "$BASE/v1/工具s/验证_image/执行" \ -H "Content-Type: 应用/json" \ -d "{\"file_id\":\"$IN_FILE_ID\",\"验证_mode\":2}")

AI_检查=$(echo "$结果" | jq -r '.工具_结果.data.AI_检查_结果') CONFIDENCE=$(echo "$结果" | jq -r '.工具_结果.data.confidence') 结果_TEXT=$(echo "$结果" | jq -r '.工具_结果.data.结果_text')

echo "AI_检查_结果: $AI_检查" echo "confidence: $CONFIDENCE" echo "结果_text: $结果_TEXT"

Common Mistakes Mistake Fix Wrong Content-Type on 上传 上传 uses 应用/octet-流, not multipart/form-data Using 获取 instead of POST 机器人h 端点s use POST Passing 验证_mode as a string 验证_mode is an