feat: 初始化 ddddocr-rs 核心推理流程
- 实现基础 OCR 流水线:图片读取 -> 灰度化 -> 张量转换 -> 推理 -> CTC 解码 - 集成 `tract-onnx` 推理引擎,实现纯 Rust 运行环境 - 完善图片预处理逻辑(基于 Lanczos3 缩放与标准化处理) - 添加 MIT 和 Apache-2.0 双重开源协议 - 完成初步的项目结构搭建与字符集(Charset)映射
This commit is contained in:
10
Cargo.toml
Normal file
10
Cargo.toml
Normal file
@@ -0,0 +1,10 @@
|
||||
[package]
|
||||
name = "ddddocr-rs"
|
||||
version = "0.1.0"
|
||||
edition = "2024"
|
||||
license = "MIT OR Apache-2.0"
|
||||
|
||||
[dependencies]
|
||||
tract-onnx = { version = "0.21.1" }
|
||||
anyhow = "1.0.102"
|
||||
image = "0.25.10"
|
||||
Reference in New Issue
Block a user