Files
ilaandClaude Sonnet 4.6 68e4f444c5 feat(vision): 实现 mingxi-vision 推理服务核心功能
- engine/defect_classes: HRIPCB 6类缺陷定义 + 4类扩展 + severity排序
- engine/base: DefectBox / DetectResult 数据类 + BaseInferenceEngine 抽象基类
- engine/ultralytics_adapter: YOLOv8 .pt 推理适配器 + warmup
- engine/annotator: 按 severity 着色的标注图生成 + base64 编码
- engine/loader: 单例管理 + asyncio.Lock 推理串行化 + 热重载支持
- config: Pydantic v1 BaseSettings,读取 .env
- schema: DetectResponse / HealthResponse / ReloadRequest 等响应模型
- api/detect: POST /api/detect,内存读图,推理锁保护
- api/health: GET /api/health
- api/model: POST /api/model/reload 热重载接口
- main: FastAPI lifespan 启动加载模型

测试通过:/api/health 200,/api/detect 空图返回 pass,标注图 base64 正常

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-05-24 21:18:34 +08:00

60 lines
1.3 KiB
Python

from abc import ABC, abstractmethod
from dataclasses import dataclass
from typing import List
import numpy as np
from .defect_classes import max_severity
@dataclass
class DefectBox:
class_id: int
class_name: str
class_name_zh: str
confidence: float
severity: str # fatal / major / minor / rework / none
box_xyxy: List[float]
@dataclass
class DetectResult:
defects: List[DefectBox]
duration_ms: float
image_width: int
image_height: int
model_version: str
@property
def defect_count(self) -> int:
return len(self.defects)
@property
def max_severity(self) -> str:
return max_severity([d.severity for d in self.defects]) if self.defects else "none"
@property
def avg_confidence(self) -> float:
if not self.defects:
return 0.0
return round(sum(d.confidence for d in self.defects) / len(self.defects), 4)
@property
def verdict(self) -> str:
return "fail" if self.defects else "pass"
class BaseInferenceEngine(ABC):
@abstractmethod
def detect(self, image: np.ndarray, conf: float = 0.45) -> DetectResult:
...
@abstractmethod
def warmup(self) -> None:
...
@property
@abstractmethod
def model_version(self) -> str:
...