diff --git a/Cargo.toml b/Cargo.toml index 9b910e5..86a4203 100644 --- a/Cargo.toml +++ b/Cargo.toml @@ -12,6 +12,8 @@ base64 = "0.22.1" imageproc = { version = "0.26.2", default-features = true } serde = { version = "1.0.228", features = ["derive"] } serde_json = "1.0.150" +ndarray="0.16.1" + [features] default = [] diff --git a/src/algo/slide.rs b/src/algo/slide.rs index 3a7e096..fc1b91c 100644 --- a/src/algo/slide.rs +++ b/src/algo/slide.rs @@ -1,24 +1,33 @@ use crate::utils::cv_ops; use crate::utils::cv_ops::{abs_diff, min_max_loc, ndarray_to_luma8, rgb_to_gray}; use crate::utils::image_io::image_to_ndarray; -use anyhow::{Context, Result, anyhow}; -use image::{DynamicImage, GenericImageView}; -use image::{ImageBuffer, Luma}; +use anyhow::{Result, anyhow}; +use image::DynamicImage; +use image::Luma; use imageproc::contrast::{ThresholdType, threshold}; use imageproc::distance_transform::Norm; use imageproc::edges::canny; use imageproc::morphology::{close, open}; use imageproc::region_labelling::{Connectivity, connected_components}; use imageproc::template_matching::{MatchTemplateMethod, match_template}; -use std::cmp::{max, min}; -use tract_onnx::prelude::tract_ndarray::{Array2, Array3, ArrayView2, ArrayView3, Axis, s}; - +use std::fmt; +use tract_onnx::prelude::tract_ndarray::{ArrayView2, ArrayView3}; +#[derive(Debug)] pub struct SlideResult { pub target: [i32; 2], pub target_x: i32, pub target_y: i32, pub confidence: f64, } +impl fmt::Display for SlideResult { + fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { + writeln!(f, "滑块匹配测试结果:")?; + writeln!(f, "检测坐标: [x: {}, y: {}]", self.target_x, self.target_y)?; + // 注意：这里保留 4 位小数，如果想让外部控制，也可以直接写 {:.4} + write!(f, "置信度: {:.4}", self.confidence)?; + Ok(()) + } +} pub struct Slider; @@ -58,23 +67,8 @@ impl Slider { target: ArrayView3, background: ArrayView3, ) -> Result { - // let (h, w, _) = target.dim(); - // 1. 计算图像差异并灰度化 (对应 cv2.absdiff + cv2.cvtColor) - // 使用 OpenCV 标准权重公式：0.299R + 0.587G + 0.114B - // let mut diff_buffer = ImageBuffer::new(w as u32, h as u32); - // for y in 0..h { - // for x in 0..w { - // let r_diff = (target[[y, x, 0]] as i16 - background[[y, x, 0]] as i16).abs() as f32; - // let g_diff = (target[[y, x, 1]] as i16 - background[[y, x, 1]] as i16).abs() as f32; - // let b_diff = (target[[y, x, 2]] as i16 - background[[y, x, 2]] as i16).abs() as f32; - // - // let gray_diff = (0.299 * r_diff + 0.587 * g_diff + 0.114 * b_diff) as u8; - // diff_buffer.put_pixel(x as u32, y as u32, Luma([gray_diff])); - // } - // } // 1. 计算差异数组 (复用 cv2::absdiff) - let (th, tw, tc) = target.dim(); let (bh, bw, bc) = background.dim(); @@ -193,9 +187,6 @@ impl Slider { background: ArrayView2, ) -> Result { // 1. 将 ndarray 转换为 imageproc 需要的 ImageBuffer (无拷贝或轻量转换) - - // let (bh, bw) = background.dim(); - // 转换逻辑 (假设你已经有方法转回 ImageBuffer) let t_buf = ndarray_to_luma8(target); let b_buf = ndarray_to_luma8(background); diff --git a/src/lib.rs b/src/lib.rs index f5ffe9c..2960071 100644 --- a/src/lib.rs +++ b/src/lib.rs @@ -3,175 +3,7 @@ mod error; pub mod models; pub mod utils; - pub use crate::algo::{SlideResult, Slider}; pub use crate::models::det::{DetBuilder, DetSession, DetectionResult, Detector}; pub use crate::models::ocr::{Ocr, OcrBuilder, OcrResult, OcrSession}; -use models::loader::ModelSession; -pub use models::ocr::model_metadata::ModelMetadata; - -// pub enum ModelSpec { -// /// 默认 OCR (使用内置路径) -// OcrModel, -// DetModel, -// /// 自定义 OCR (路径由用户提供) -// CustomOcrModel { -// path: String, -// model_metadata: ModelMetadata, -// }, -// } -// impl ModelSpec { -// // 将默认路径定义为内部关联常量 -// const DEFAULT_OCR_PATH: &'static str = "models/common_sml2h3_f32.onnx"; -// const DEFAULT_DET_PATH: &'static str = "models/common_det.onnx"; -// } -// pub enum Runtime { -// Ocr(Ocr), -// Det(Det), -// } -// impl Runtime { -// // 统一获取描述的方法 -// pub fn desc(&self) -> String { -// match self { -// Runtime::Ocr(s) => s.desc(), // 调用 Ocr 结构体的方法 -// Runtime::Det(s) => s.desc(), // 调用 Det 结构体的方法 -// } -// } -// } -// pub struct DdddOcrBuilder { -// mode: ModelSpec, -// } -// -// impl DdddOcrBuilder { -// pub fn new() -> Self { -// Self { -// mode: ModelSpec::OcrModel, -// } -// } -// -// /// 切换为检测模式 -// pub fn det(mut self) -> Self { -// self.mode = ModelSpec::DetModel; -// self -// } -// -// /// 设置自定义 OCR 路径 -// pub fn custom_ocr(mut self, path: String, model_metadata: ModelMetadata) -> Self { -// // 直接重写枚举，替换掉之前的 Ocr 或 Det -// self.mode = ModelSpec::CustomOcrModel { -// path, -// model_metadata, -// }; -// self -// } -// -// /// 核心初始化逻辑 -// pub fn build(self) -> Result { -// let runtime = match self.mode { -// ModelSpec::OcrModel => Runtime::Ocr(Ocr::new( -// ModelSpec::DEFAULT_OCR_PATH.into(), -// ModelMetadata::from_builtin_beta(), -// )?), -// ModelSpec::DetModel => Runtime::Det(Det::new(ModelSpec::DEFAULT_DET_PATH.into())?), -// ModelSpec::CustomOcrModel { -// path, -// model_metadata, -// } => Runtime::Ocr(Ocr::new(path, model_metadata)?), -// }; -// -// Ok(DdddOcr { runtime }) -// } -// } -// -// pub struct DdddOcr { -// runtime: Runtime, -// } -// -// impl Display for DdddOcr { -// fn fmt(&self, f: &mut Formatter<'_>) -> std::fmt::Result { -// write!(f, "DdddOcr(ocr: {})", self.runtime.desc()) -// } -// } -// -// impl DdddOcr { -// pub fn classification(&self, img: &DynamicImage) -> Result { -// match &self.runtime { -// // Runtime::Ocr(s) => s.predict(img).run(), -// // Runtime::Ocr(s) => s.predictor().probability(false).predict(img), -// // Runtime::Ocr(s) => { -// // let predictor = s.predictor(); -// // let restricted = predictor.charset_restrict(&CharRestrict::Lowercase); -// // let a = restricted.valid_tokens(); -// // println!("{:?}", a); -// // Ok("".to_string()) -// // } -// Runtime::Ocr(s) => { -// let res = s.predictor().probability(true).predict(img)?; -// println!("{}", res); -// Ok(res.to_string()) -// } -// // Runtime::Ocr(s) => s.predictor().charset_restrict(&CharRestrict::Digit).predict(img), -// // Runtime::Ocr(s) => s.predictor().color_filter(&ColorPreset::Custom(vec![ -// // // 错误：下界 (82, 221, 14) 没问题 -// // // 但上界的 H 通道写成了 240，超过了 180 的法定上限！ -// // HsvRange::new((82, 221, 14), (240, 203, 82)), -// // ])).predict(img), -// Runtime::Det(_) => Err(anyhow::anyhow!("当前模型是检测模型，无法执行 OCR")), -// } -// } -// pub fn detection(&self, img: &DynamicImage) -> Result> { -// match &self.runtime { -// Runtime::Det(s) => s.predict(img), -// Runtime::Ocr(_) => Err(anyhow::anyhow!("当前模型是 OCR 模型，无法执行检测")), -// } -// } -// } - -// struct Classification {} -// #[derive(Debug)] -// struct ClassificationBuilder { -// img: DynamicImage, -// png_fix: bool, -// color_filter_colors: Option>, -// color_filter_custom_ranges: Option>, -// } -// impl ClassificationBuilder { -// pub fn new(img: DynamicImage) -> Self { -// ClassificationBuilder { -// img, -// png_fix: false, -// color_filter_colors: None, -// color_filter_custom_ranges: None, -// } -// } -// pub fn png_fix(mut self, value: bool) -> Self { -// self.png_fix = value; -// self -// } -// pub fn color_filter_colors(mut self, value: Vec) -> Self { -// self.color_filter_colors = Some(value); -// self -// } -// pub fn color_filter_custom_ranges(mut self, value: Vec) -> Self { -// self.color_filter_custom_ranges = Some(value); -// self -// } -// pub fn build(self) -> Classification { -// Classification {} -// } -// } - -#[cfg(test)] -mod tests { - #[test] - fn test_ctc_decode_indices() { - // 模拟一个 DdddOcr 实例（如果 decode 不依赖 ocr，可以设为相关函数） - // 这里假设你的 decode_ctc 是公开或内部可访问的 - let input = vec![1, 1, 0, 1, 2, 2, 0, 2]; - // 逻辑：[1, 1] -> 1, [0] -> 跳过, [1] -> 1, [2, 2] -> 2, [0] -> 跳过, [2] -> 2 - // 预期结果索引应该是 [1, 1, 2, 2] 对应的字符 - // 具体的断言取决于你的 CHARSET_BETA - // let result = dddd.ctc_decode_indices(&input); - // assert_eq!(result, "AABB"); - } -} +pub use models::ocr::metadata::ModelMetadata; diff --git a/src/models/det/executor.rs b/src/models/det/executor.rs index 6cbb8b9..ce95591 100644 --- a/src/models/det/executor.rs +++ b/src/models/det/executor.rs @@ -1,15 +1,10 @@ -use crate::models::ocr::model_metadata::ModelMetadata; -use crate::models::loader::{ModelLoader, ModelSession, ModelType}; -use anyhow::{Context, Result, anyhow}; -use image::{DynamicImage, GenericImageView, imageops::FilterType}; -use std::path::Path; -use tract_onnx::prelude::tract_ndarray::{Array2, Array3, Array4, Axis, prelude::*, s}; -use tract_onnx::prelude::{Graph, RunnableModel, Tensor, TypedFact, TypedOp, tvec}; +use anyhow::{Context, Result}; +use image::{imageops::FilterType, DynamicImage, GenericImageView}; +use std::fmt; +use tract_onnx::prelude::tract_ndarray::{prelude::*, s, Array2, Array3, Array4, Axis}; +use tract_onnx::prelude::{Tensor}; -const DEFAULT_DET_PATH: &'static str = "common_det.onnx"; -// 预设的提示信息常量 -use crate::error::MODEL_DOWNLOAD_HELP; use crate::models::det::session::DetSession; #[derive(Debug, Clone, Copy)] @@ -22,15 +17,23 @@ pub struct DetectionResult { pub class_id: u32, } - - - - +impl fmt::Display for DetectionResult { + fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { + // 结构体只管自己这一行怎么显示，不用管外部的索引 [i] + write!( + f, + "x1={}, y1={}, x2={}, y2={}, 分数={:.4}, 类别ID={}", + self.x1, self.y1, self.x2, self.y2, self.score, self.class_id + ) + } +} #[derive(Debug)] pub struct Detector<'a> { pub(crate) session: &'a DetSession, + #[allow(dead_code)] pub(crate) use_gpu: bool, + #[allow(dead_code)] pub(crate) device_id: u8, } diff --git a/src/models/loader.rs b/src/models/loader.rs index 32eedd9..75b425d 100644 --- a/src/models/loader.rs +++ b/src/models/loader.rs @@ -1,17 +1,10 @@ -use anyhow::{anyhow, Context}; -use image::DynamicImage; +use anyhow::Context; +use std::io::Cursor; use tract_onnx::onnx; use tract_onnx::prelude::*; -// 关键点：直接使用 tract 重导出的 ndarray -use crate::utils::image_io::png_rgba_white_preprocess; -use crate::utils::image_processor::{convert_to_grayscale, resize_image}; -use std::collections::HashMap; -use std::io::Cursor; -use tract_onnx::prelude::tract_ndarray::s; -use crate::ModelMetadata; /// OCR 模型：包含路径和字符集 -const DEFAULT_OCR_PATH: &'static str = "common_sml2h3_f32.onnx"; + pub enum ModelType { Ocr, Det, @@ -53,60 +46,3 @@ impl ModelLoader { Ok(Self { session }) } } -// impl ModelLoader { -// pub fn find_model_path(env_var: &str, default_filename: &str) -> Option { -// // 1. 策略一：优先尝试读取环境变量 -// if let Ok(env_path) = std::env::var(env_var) { -// let path = std::path::PathBuf::from(env_path); -// if path.exists() { -// return Some(path); -// } -// } -// // 2. 策略二：尝试在当前工作目录寻找 -// if let Ok(mut path) = std::env::current_dir() { -// path.push(default_filename); -// if path.exists() { -// return Some(path); -// } -// } -// -// // 3. 策略三：尝试在当前可执行文件同级目录寻找 -// if let Ok(mut exe_path) = std::env::current_exe() { -// exe_path.pop(); // 弹出可执行文件名，拿到所在的父目录 -// exe_path.push(default_filename); -// if exe_path.exists() { -// return Some(exe_path); -// } -// } -// // 4. 所有本地探测策略均落空 -// None -// } -// } - - - -// pub fn new_default() -> anyhow::Result { -// let metadata = ModelMetadata::from_builtin_beta(); // 绑定自带的 BETA 字符集 -// -// // 1. 策略一：优先尝试读取环境变量 -// if let Some(path) = ModelLoader::find_model_path("DDDD_OCR_MODEL", DEFAULT_OCR_PATH) { -// return Self::new(path, metadata); -// } -// -// // 4. 策略四：开启了 embed-models 特征时的编译期死穴保底 -// // 如果开启了 feature 但根目录下没有该模型，编译时会在此处直接中断失败 -// -// #[cfg(feature = "embed-models")] -// { -// let model_bytes = include_bytes!("../models/common_sml2h3_f32.onnx"); -// // 注意：这里需要你的 InternalOcr 扩展一个 from_bytes 的方法 -// return Self::model_from_bytes(model_bytes, metadata); -// -// } -// -// // 5. 所有策略落空，抛出保姆级错误 -// #[cfg(not(feature = "embed-models"))] -// { -// Err(anyhow!(MODEL_DOWNLOAD_HELP)) -// } -// } \ No newline at end of file diff --git a/src/models/ocr/builder.rs b/src/models/ocr/builder.rs index 20b07ec..20087ea 100644 --- a/src/models/ocr/builder.rs +++ b/src/models/ocr/builder.rs @@ -1,7 +1,8 @@ -use crate::models::ocr::charset::TokenFilter; use crate::models::ocr::executor::Ocr; use crate::models::ocr::session::OcrSession; use crate::models::ocr::color_filter::ColorFilter; +use crate::models::ocr::token_filter::TokenFilter; + pub struct OcrBuilder { /// 是否修复PNG格式问题 png_fix: bool, diff --git a/src/models/ocr/color_filter.rs b/src/models/ocr/color_filter.rs index 24bb808..0b97e89 100644 --- a/src/models/ocr/color_filter.rs +++ b/src/models/ocr/color_filter.rs @@ -1,25 +1,24 @@ -use std::str::FromStr; +use crate::utils::cv_ops::rgb_to_opencv_hsv; use anyhow::anyhow; use image::{DynamicImage, ImageBuffer, Rgb}; -use crate::utils::cv_ops::rgb_to_opencv_hsv; - -#[derive(Debug, Clone, Copy, PartialEq, Eq, PartialOrd, Ord)] -pub struct HsvRange { - pub lower: (u8, u8, u8), // (H, S, V) - pub upper: (u8, u8, u8), // (H, S, V) -} - +use std::str::FromStr; /// 核心区间判定辅助函数 #[inline(always)] fn is_pixel_matched(ranges: &[HsvRange], h: u8, s: u8, v: u8) -> bool { ranges.iter().any(|range| { - h >= range.lower.0 && h <= range.upper.0 && - s >= range.lower.1 && s <= range.upper.1 && - v >= range.lower.2 && v <= range.upper.2 + h >= range.lower.0 + && h <= range.upper.0 + && s >= range.lower.1 + && s <= range.upper.1 + && v >= range.lower.2 + && v <= range.upper.2 }) } -pub fn filter_image(image: &DynamicImage, hsv_ranges: &[HsvRange]) -> anyhow::Result { +pub fn apply_to_image( + image: &DynamicImage, + hsv_ranges: &[HsvRange], +) -> anyhow::Result { // 1. 统一转换为连续内存的 RGB8 缓冲区 (对应 Python 的 Image 到 RGB/BGR 数组转换) let rgb_img = image.to_rgb8(); let (width, height) = rgb_img.dimensions(); @@ -50,6 +49,13 @@ pub fn filter_image(image: &DynamicImage, hsv_ranges: &[HsvRange]) -> anyhow::Re Ok(DynamicImage::ImageRgb8(filtered_buffer)) } + +#[derive(Debug, Clone, Copy, PartialEq, Eq, PartialOrd, Ord)] +pub struct HsvRange { + pub lower: (u8, u8, u8), // (H, S, V) + pub upper: (u8, u8, u8), // (H, S, V) +} + impl HsvRange { pub const fn new(lower: (u8, u8, u8), upper: (u8, u8, u8)) -> Self { Self { lower, upper } @@ -65,7 +71,8 @@ impl HsvRange { } // 2. 校验下界不能大于上界 - if self.lower.0 > self.upper.0 || self.lower.1 > self.upper.1 || self.lower.2 > self.upper.2 { + if self.lower.0 > self.upper.0 || self.lower.1 > self.upper.1 || self.lower.2 > self.upper.2 + { return Err("HSV范围下界不能大于上界".to_string()); } @@ -87,25 +94,58 @@ pub enum ColorPreset { Custom(Vec), } -impl ColorPreset { +impl ColorPreset { /// 纯裸数据定义，没有任何结构体包装，干净利落 /// 返回值：(范围数量, 范围数组) /// 完美的零成本抽象：利用常量提升将数据直接打入只读数据段 (.rodata) pub fn matches(&self) -> &[HsvRange] { match self { ColorPreset::Red => &[ - HsvRange { lower: (0, 50, 50), upper: (10, 255, 255) }, - HsvRange { lower: (170, 50, 50), upper: (180, 255, 255) }, + HsvRange { + lower: (0, 50, 50), + upper: (10, 255, 255), + }, + HsvRange { + lower: (170, 50, 50), + upper: (180, 255, 255), + }, ], - ColorPreset::Blue => &[HsvRange { lower: (100, 50, 50), upper: (130, 255, 255) }], - ColorPreset::Green => &[HsvRange { lower: (40, 50, 50), upper: (80, 255, 255) }], - ColorPreset::Yellow => &[HsvRange { lower: (20, 50, 50), upper: (40, 255, 255) }], - ColorPreset::Orange => &[HsvRange { lower: (10, 50, 50), upper: (20, 255, 255) }], - ColorPreset::Purple => &[HsvRange { lower: (130, 50, 50), upper: (170, 255, 255) }], - ColorPreset::Cyan => &[HsvRange { lower: (80, 50, 50), upper: (100, 255, 255) }], - ColorPreset::Black => &[HsvRange { lower: (0, 0, 0), upper: (180, 255, 50) }], - ColorPreset::White => &[HsvRange { lower: (0, 0, 200), upper: (180, 30, 255) }], - ColorPreset::Gray => &[HsvRange { lower: (0, 0, 50), upper: (180, 30, 200) }], + ColorPreset::Blue => &[HsvRange { + lower: (100, 50, 50), + upper: (130, 255, 255), + }], + ColorPreset::Green => &[HsvRange { + lower: (40, 50, 50), + upper: (80, 255, 255), + }], + ColorPreset::Yellow => &[HsvRange { + lower: (20, 50, 50), + upper: (40, 255, 255), + }], + ColorPreset::Orange => &[HsvRange { + lower: (10, 50, 50), + upper: (20, 255, 255), + }], + ColorPreset::Purple => &[HsvRange { + lower: (130, 50, 50), + upper: (170, 255, 255), + }], + ColorPreset::Cyan => &[HsvRange { + lower: (80, 50, 50), + upper: (100, 255, 255), + }], + ColorPreset::Black => &[HsvRange { + lower: (0, 0, 0), + upper: (180, 255, 50), + }], + ColorPreset::White => &[HsvRange { + lower: (0, 0, 200), + upper: (180, 30, 255), + }], + ColorPreset::Gray => &[HsvRange { + lower: (0, 0, 50), + upper: (180, 30, 200), + }], ColorPreset::Custom(ranges) => ranges, } } @@ -204,7 +244,6 @@ impl ColorFilter for ColorPreset { } } - /// 多路颜色“或”逻辑组合子（并集网络） pub struct MultiOrColorRestrict<'a> { pub filters: Vec<&'a dyn ColorFilter>, @@ -246,4 +285,3 @@ macro_rules! color_any_of { } }; } - diff --git a/src/models/ocr/executor.rs b/src/models/ocr/executor.rs index 8ce4d4b..05a72bf 100644 --- a/src/models/ocr/executor.rs +++ b/src/models/ocr/executor.rs @@ -1,7 +1,7 @@ -use crate::models::ocr::model_metadata::Resize; +use crate::models::ocr::metadata::Resize; use crate::models::ocr::session::OcrSession; -use crate::models::ocr::color_filter::{HsvRange, filter_image}; +use crate::models::ocr::color_filter::{HsvRange, apply_to_image}; use crate::utils::image_io::png_rgba_white_preprocess; use crate::utils::image_processor::{convert_to_grayscale, resize_image}; use anyhow::Result; @@ -141,7 +141,7 @@ impl<'a> Ocr<'a> { } Ok(Some(ranges)) => { // 只有真正需要过滤时，才在内部提取像素并生成清洗后的 Owned 新图 - let filtered_img = filter_image(image, ranges)?; + let filtered_img = apply_to_image(image, ranges)?; Cow::Owned(filtered_img) } }; diff --git a/src/models/ocr/model_metadata.rs b/src/models/ocr/metadata.rs similarity index 64% rename from src/models/ocr/model_metadata.rs rename to src/models/ocr/metadata.rs index 63aef00..cc4e61f 100644 --- a/src/models/ocr/model_metadata.rs +++ b/src/models/ocr/metadata.rs @@ -1,11 +1,72 @@ -use crate::models::ocr::charset::{CHARSET_BETA, CHARSET_OLD, Charset}; -use anyhow::{Result, anyhow}; +use anyhow::{anyhow, Result}; use serde::Deserialize; use std::borrow::Cow; -use std::collections::{HashMap, HashSet}; -use std::fs::File; -use std::io::Read; -use std::path::Path; +use std::collections::HashMap; + +// ========================================== +// 3. 字符集核心结构体 (重命名为 Charset) +// ========================================== +#[derive(Debug, Clone)] +pub struct Charset { + // 使用 Cow 统一静态切片和动态读取的 Vec，内部实现真正的零拷贝 + pub tokens: Vec>, + // 反向查找表，保证字符转索引为 O(1) + pub char_to_idx: HashMap, usize>, + // 当前处于激活状态的有效索引缓存 (用于 CTC 解码前的过滤加速) + // pub valid_indices: HashSet, +} + +impl Charset { + // 内部底层统一收拢构造 + pub fn new(tokens: Vec>) -> Self { + let mut char_to_idx = HashMap::with_capacity(tokens.len()); + for (idx, token) in tokens.iter().enumerate() { + char_to_idx.entry(token.clone()).or_insert(idx); + // 如果字符集有重复，保留第一个遇到的索引 (符合 Python .index 逻辑) + // char_to_idx.entry(token.to_string()).or_insert(idx); + } + + Self { + tokens, + char_to_idx, + } + } + + // --- 业务策略方法 --- + + /// 将字符转为索引，不存在返回 -1 (保持与原 Python 库行为一致) + pub fn char_to_index(&self, char_str: &str) -> i32 { + if let Some(&idx) = self.char_to_idx.get(char_str) { + idx as i32 + } else { + -1 + } + } + + /// 将索引转为字符引用，零拷贝。若越界返回 None + pub fn index_to_char_ref(&self, index: usize) -> Option<&str> { + self.tokens.get(index).map(|cow| cow.as_ref()) + } + + pub fn is_valid_char(&self, char_str: &str) -> bool { + self.char_to_idx.get(char_str).is_some() + } + pub fn size(&self) -> usize { + self.tokens.len() + } +} + +// ========================================== +// 4. 标准 Display 接口实现 (对应 __str__) +// ========================================== +impl std::fmt::Display for Charset { + fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result { + write!(f, "Charset [Total Size: {}", self.size(),) + } +} + + + // ===================================================================== // 1. 辅助定义的枚举与结构体 // ===================================================================== @@ -74,28 +135,6 @@ pub struct ModelMetadata { impl ModelMetadata { // --- 优雅的工厂模式构造器 --- - /// 从预设的旧版字符集创建 - pub fn from_builtin_old() -> Self { - Self::from_static_slice( - CHARSET_OLD, - false, - Resize::DynamicWidth(64), - 1, - Normalization::ZeroToOne, - ) - } - - /// 从预设的 Beta 版字符集创建 - pub fn from_builtin_beta() -> Self { - Self::from_static_slice( - CHARSET_BETA, - false, - Resize::DynamicWidth(64), - 1, - Normalization::MinusOneToOne, - ) - } - /// 通用的静态切片转换构造器 pub fn from_static_slice( slice: &[&'static str], @@ -158,54 +197,4 @@ impl ModelMetadata { .map_err(|e| anyhow!("JSON 字节流不是合法的 UTF-8 编码: {}", e))?; Self::from_json_str(json_str) } - - /// 从外部外部 JSON 文件动态加载字符集(在后续优化中移除) - pub fn from_json_file>(path: P) -> Result { - let path = path.as_ref(); - if !path.exists() { - return Err(anyhow!("模型元数据配置文件不存在: {:?}", path)); - } - - let mut file = File::open(path)?; - let mut content = String::new(); - file.read_to_string(&mut content)?; - - let dto: ModelMetadataDto = serde_json::from_str(&content) - .map_err(|e| anyhow!("JSON 反序列化失败，请检查字段是否完整: {}", e))?; - - // 1. 将 DTO 的字符串数组转化为强类型的 Charset - let tokens: Vec> = - dto.charset.into_iter().map(|s| Cow::Owned(s)).collect(); - let charset = Charset::new(tokens); - - // 2. 解析 resize 策略（重现 Python 的复杂条件判断） - if dto.resize.len() != 2 { - return Err(anyhow!( - "'resize (or image)' 字段必须是包含两个元素的数组，例如 [-1, 64]" - )); - } - let r0 = dto.resize[0]; - let r1 = dto.resize[1]; - - let resize = if r0 == -1 { - if dto.word { - // 如果 word 为 true，且包含 -1，Python 里是 resize 为 (r1, r1) 的正方形 - Resize::Square(r1 as u32) - } else { - // 如果 word 为 false，且包含 -1，Python 里是高度固定为 r1，宽度按原图比例缩放 - Resize::DynamicWidth(r1 as u32) - } - } else { - // 正常的固定宽高 - Resize::Fixed(r0 as u32, r1 as u32) - }; - - Ok(Self { - charset, - word: dto.word, - resize, - channel: dto.channel, - normalization: dto.normalization, - }) - } } diff --git a/src/models/ocr/mod.rs b/src/models/ocr/mod.rs index ecf7fac..63be5ad 100644 --- a/src/models/ocr/mod.rs +++ b/src/models/ocr/mod.rs @@ -1,9 +1,9 @@ mod builder; mod executor; mod session; -pub mod charset; -pub mod model_metadata; +pub mod metadata; pub mod color_filter; +mod token_filter; pub use builder::OcrBuilder; pub use executor::{Ocr, OcrResult}; diff --git a/src/models/ocr/session.rs b/src/models/ocr/session.rs index 4bd056d..4a226af 100644 --- a/src/models/ocr/session.rs +++ b/src/models/ocr/session.rs @@ -1,4 +1,4 @@ -use crate::models::ocr::model_metadata::ModelMetadata; +use crate::models::ocr::metadata::ModelMetadata; use crate::models::loader::{ModelLoader, ModelSession, ModelType}; use anyhow::Context; use anyhow::Result; diff --git a/src/models/ocr/token_filter.rs b/src/models/ocr/token_filter.rs new file mode 100644 index 0000000..dd5f58c --- /dev/null +++ b/src/models/ocr/token_filter.rs @@ -0,0 +1,146 @@ +use std::borrow::Cow; + +/// 字符集范围限制枚举 +pub struct ValidationCtx<'a> { + pub text: &'a str, // 当前 Token 的文本内容 + pub token_id: usize, // 当前 Token 的 ID 索引 +} + +/// 统一的约束接口 +pub trait TokenFilter { + fn matches(&self, ctx: &ValidationCtx) -> bool; + /// 预估容量提示，帮助精准开辟 Vec 内存 + fn estimated_capacity(&self) -> usize { + 128 + } + /// 【新引入的架构级核心方法】 + /// 统一接管全量字符集的密集遍历、CTC Blank放行、去重、排序及空交集退化兜底 + fn apply_to_charset(&self, tokens: &[Cow]) -> Option> { + let mut has_any_match = false; + let estimated_capacity = self.estimated_capacity(); + + // 1. 精准开辟内存，完美利用容量提示，避免动态乱涨 + let mut temp_indices = Vec::with_capacity(estimated_capacity.max(16)); + + // 2. 高性能原地单次流式迭代 + for (idx, token) in tokens.iter().enumerate() { + let token_str = token.as_ref(); + + // 规则 A: CTC Blank 空字符串或 0 号索引无条件放行 + if token_str.is_empty() || idx == 0 { + temp_indices.push(idx); + continue; // 关键：直接跳过，防止后续 matches 匹配成功导致重复 push 产生 Bug + } + + // 规则 B: 组装无拷贝上下文 + let ctx = ValidationCtx { + text: token_str, + token_id: idx, + }; + + // 规则 C: 路由到各自具体实现的特异性匹配中（如 Digit 判定、TopN 判定、组合子判定等） + if self.matches(&ctx) { + temp_indices.push(idx); + has_any_match = true; + } + } + + // 3. 终极防御：如果整个模型字符集除了 Blank，一个都没对上，直接退化为 None（全量识别） + if !has_any_match { + println!("警告：当前限制策略与模型字符集完全没有交集！已自动恢复全量识别。"); + None + } else { + // 4. 排序并去重，为 Ocr 引擎后续进行极其高频的『二分查找』筑起绝对安全的底层保障 + temp_indices.sort_unstable(); + temp_indices.dedup(); + Some(temp_indices) + } + } +} + +#[derive(Debug, Clone, PartialEq, Eq)] +pub enum CharRestrict { + Digit, + Lowercase, + Uppercase, + CustomList(Vec), +} + +impl TokenFilter for CharRestrict { + fn matches(&self, ctx: &ValidationCtx) -> bool { + match self { + Self::Digit => ctx.text.len() == 1 && ctx.text.as_bytes()[0].is_ascii_digit(), + Self::Lowercase => ctx.text.len() == 1 && ctx.text.as_bytes()[0].is_ascii_lowercase(), + Self::Uppercase => ctx.text.len() == 1 && ctx.text.as_bytes()[0].is_ascii_uppercase(), + Self::CustomList(vec) => vec.iter().any(|t| t == ctx.text), + } + } + fn estimated_capacity(&self) -> usize { + match self { + Self::Digit => 16, + Self::Lowercase | Self::Uppercase => 32, + Self::CustomList(vec) => vec.len() + 1, + } + } +} + +#[derive(Debug, Clone, PartialEq, Eq)] +pub enum IdRestrict { + TopN(usize), + IdRange(std::ops::Range), + IdList(Vec), +} + +impl TokenFilter for IdRestrict { + fn matches(&self, ctx: &ValidationCtx) -> bool { + match self { + Self::TopN(n) => ctx.token_id < *n, + Self::IdRange(range) => range.contains(&ctx.token_id), + Self::IdList(vec) => vec.contains(&ctx.token_id), + } + } + fn estimated_capacity(&self) -> usize { + match self { + Self::TopN(n) => *n + 1, + // 2. IdRange：标准标准库 Range 的长度 + // 注意：因为范围可能是 1000..2000，它的 len() 返回的是 usize + Self::IdRange(range) => range.len() + 1, + // 3. IdList：Vec 里的元素个数 + Self::IdList(vec) => vec.len() + 1, + } + } +} + +/// 多路“或”逻辑组合子（支持 N 个规则无缝并集） +pub struct MultiOrRestrict<'a> { + pub filters: Vec<&'a dyn TokenFilter>, +} + +impl<'a> TokenFilter for MultiOrRestrict<'a> { + fn matches(&self, ctx: &ValidationCtx) -> bool { + // 核心高阶函数：只要有一个过滤器命中，该 Token 即可放行 + self.filters.iter().any(|f| f.matches(ctx)) + } + + fn estimated_capacity(&self) -> usize { + // 将所有过滤器的预估容量累加，作为最终容量参考 + self.filters.iter().map(|f| f.estimated_capacity()).sum() + } +} +// ===================================================================== +// 声明式宏：替代 `+` 运算符，解决组合扩展痛苦 +// ===================================================================== +#[macro_export] +macro_rules! any_of { + // 场景 A：如果用户只传了一个规则，免去构建 Vec 的开销，直接返回其引用 + ($only:expr) => { + &$only as &dyn $crate::TokenFilter + }; + + // 场景 B：如果用户传入了多个规则，自动织成一张静态组合网 + ($($filter:expr),+ $(,)?) => { + &$crate::MultiOrRestrict { + filters: vec![ $( &$filter as &dyn $crate::TokenFilter ),+ ] + } + }; +} diff --git a/src/utils/image_processor.rs b/src/utils/image_processor.rs index c925277..6531aff 100644 --- a/src/utils/image_processor.rs +++ b/src/utils/image_processor.rs @@ -1,5 +1,7 @@ -use image::{DynamicImage, GrayImage, imageops::FilterType}; -use anyhow::Result; +use image::{DynamicImage, GrayImage, imageops::FilterType, Rgb, ImageBuffer}; +use anyhow::{anyhow, Result}; +use crate::models::ocr::color_filter::HsvRange; +use crate::utils::cv_ops::rgb_to_opencv_hsv; /// 对应 Python 的 convert_to_grayscale /// 将图像转换为灰度图 (L模式) @@ -34,4 +36,5 @@ pub fn resize_image( // target_height, // FilterType::Lanczos3 // ) -// } \ No newline at end of file +// } + diff --git a/src/models/ocr/charset.rs b/tests/char_slice.rs similarity index 88% rename from src/models/ocr/charset.rs rename to tests/char_slice.rs index c9de064..59beb89 100644 --- a/src/models/ocr/charset.rs +++ b/tests/char_slice.rs @@ -1,3 +1,10 @@ +use std::borrow::Cow; +use std::fs::File; +use std::path::Path; +use anyhow::anyhow; +use ddddocr_rs::models::ocr::metadata::Charset; +use ddddocr_rs::models::ocr::metadata::{Normalization, Resize}; + pub const CHARSET_BETA: &[&str] = &[ "", "笤", "谴", "膀", "荔", "佰", "电", "臁", "矍", "同", "奇", "芄", "吠", "6", "曛", "荇", "砥", "蹅", "晃", "厄", "殣", "ｃ", "辱", "钋", "杻", "價", "眙", "鴿", "⒄", "裙", "训", "涛", @@ -517,212 +524,77 @@ pub const CHARSET_BETA: &[&str] = &[ pub const CHARSET_OLD: &[&str] = &["", "笤", "谴", "膀", "荔"]; -use std::borrow::Cow; -use std::collections::{HashMap, HashSet}; -/// 字符集范围限制枚举 -pub struct ValidationCtx<'a> { - pub text: &'a str, // 当前 Token 的文本内容 - pub token_id: usize, // 当前 Token 的 ID 索引 -} -/// 统一的约束接口 -pub trait TokenFilter { - fn matches(&self, ctx: &ValidationCtx) -> bool; - /// 预估容量提示，帮助精准开辟 Vec 内存 - fn estimated_capacity(&self) -> usize { - 128 - } - /// 【新引入的架构级核心方法】 - /// 统一接管全量字符集的密集遍历、CTC Blank放行、去重、排序及空交集退化兜底 - fn apply_to_charset(&self, tokens: &[Cow]) -> Option> { - let mut has_any_match = false; - let estimated_capacity = self.estimated_capacity(); - // 1. 精准开辟内存，完美利用容量提示，避免动态乱涨 - let mut temp_indices = Vec::with_capacity(estimated_capacity.max(16)); +// pub fn from_builtin_old() -> Self { +// Self::from_static_slice( +// CHARSET_OLD, +// false, +// Resize::DynamicWidth(64), +// 1, +// Normalization::ZeroToOne, +// ) +// } +// +// /// 从预设的 Beta 版字符集创建 +// pub fn from_builtin_beta() -> Self { +// Self::from_static_slice( +// CHARSET_BETA, +// false, +// Resize::DynamicWidth(64), +// 1, +// Normalization::MinusOneToOne, +// ) +// } - // 2. 高性能原地单次流式迭代 - for (idx, token) in tokens.iter().enumerate() { - let token_str = token.as_ref(); - // 规则 A: CTC Blank 空字符串或 0 号索引无条件放行 - if token_str.is_empty() || idx == 0 { - temp_indices.push(idx); - continue; // 关键：直接跳过，防止后续 matches 匹配成功导致重复 push 产生 Bug - } - - // 规则 B: 组装无拷贝上下文 - let ctx = ValidationCtx { - text: token_str, - token_id: idx, - }; - - // 规则 C: 路由到各自具体实现的特异性匹配中（如 Digit 判定、TopN 判定、组合子判定等） - if self.matches(&ctx) { - temp_indices.push(idx); - has_any_match = true; - } - } - - // 3. 终极防御：如果整个模型字符集除了 Blank，一个都没对上，直接退化为 None（全量识别） - if !has_any_match { - println!("警告：当前限制策略与模型字符集完全没有交集！已自动恢复全量识别。"); - None - } else { - // 4. 排序并去重，为 Ocr 引擎后续进行极其高频的『二分查找』筑起绝对安全的底层保障 - temp_indices.sort_unstable(); - temp_indices.dedup(); - Some(temp_indices) - } - } -} - -#[derive(Debug, Clone, PartialEq, Eq)] -pub enum CharRestrict { - Digit, - Lowercase, - Uppercase, - CustomList(Vec), -} - -impl TokenFilter for CharRestrict { - fn matches(&self, ctx: &ValidationCtx) -> bool { - match self { - Self::Digit => ctx.text.len() == 1 && ctx.text.as_bytes()[0].is_ascii_digit(), - Self::Lowercase => ctx.text.len() == 1 && ctx.text.as_bytes()[0].is_ascii_lowercase(), - Self::Uppercase => ctx.text.len() == 1 && ctx.text.as_bytes()[0].is_ascii_uppercase(), - Self::CustomList(vec) => vec.iter().any(|t| t == ctx.text), - } - } - fn estimated_capacity(&self) -> usize { - match self { - Self::Digit => 16, - Self::Lowercase | Self::Uppercase => 32, - Self::CustomList(vec) => vec.len() + 1, - } - } -} - -#[derive(Debug, Clone, PartialEq, Eq)] -pub enum IdRestrict { - TopN(usize), - IdRange(std::ops::Range), - IdList(Vec), -} - -impl TokenFilter for IdRestrict { - fn matches(&self, ctx: &ValidationCtx) -> bool { - match self { - Self::TopN(n) => ctx.token_id < *n, - Self::IdRange(range) => range.contains(&ctx.token_id), - Self::IdList(vec) => vec.contains(&ctx.token_id), - } - } - fn estimated_capacity(&self) -> usize { - match self { - Self::TopN(n) => *n + 1, - // 2. IdRange：标准标准库 Range 的长度 - // 注意：因为范围可能是 1000..2000，它的 len() 返回的是 usize - Self::IdRange(range) => range.len() + 1, - // 3. IdList：Vec 里的元素个数 - Self::IdList(vec) => vec.len() + 1, - } - } -} - -/// 多路“或”逻辑组合子（支持 N 个规则无缝并集） -pub struct MultiOrRestrict<'a> { - pub filters: Vec<&'a dyn TokenFilter>, -} - -impl<'a> TokenFilter for MultiOrRestrict<'a> { - fn matches(&self, ctx: &ValidationCtx) -> bool { - // 核心高阶函数：只要有一个过滤器命中，该 Token 即可放行 - self.filters.iter().any(|f| f.matches(ctx)) - } - - fn estimated_capacity(&self) -> usize { - // 将所有过滤器的预估容量累加，作为最终容量参考 - self.filters.iter().map(|f| f.estimated_capacity()).sum() - } -} -// ===================================================================== -// 声明式宏：替代 `+` 运算符，解决组合扩展痛苦 -// ===================================================================== -#[macro_export] -macro_rules! any_of { - // 场景 A：如果用户只传了一个规则，免去构建 Vec 的开销，直接返回其引用 - ($only:expr) => { - &$only as &dyn $crate::TokenFilter - }; - - // 场景 B：如果用户传入了多个规则，自动织成一张静态组合网 - ($($filter:expr),+ $(,)?) => { - &$crate::MultiOrRestrict { - filters: vec![ $( &$filter as &dyn $crate::TokenFilter ),+ ] - } - }; -} - -// ========================================== -// 3. 字符集核心结构体 (重命名为 Charset) -// ========================================== -#[derive(Debug, Clone)] -pub struct Charset { - // 使用 Cow 统一静态切片和动态读取的 Vec，内部实现真正的零拷贝 - pub tokens: Vec>, - // 反向查找表，保证字符转索引为 O(1) - pub char_to_idx: HashMap, usize>, - // 当前处于激活状态的有效索引缓存 (用于 CTC 解码前的过滤加速) - // pub valid_indices: HashSet, -} - -impl Charset { - // 内部底层统一收拢构造 - pub fn new(tokens: Vec>) -> Self { - let mut char_to_idx = HashMap::with_capacity(tokens.len()); - for (idx, token) in tokens.iter().enumerate() { - char_to_idx.entry(token.clone()).or_insert(idx); - // 如果字符集有重复，保留第一个遇到的索引 (符合 Python .index 逻辑) - // char_to_idx.entry(token.to_string()).or_insert(idx); - } - - Self { - tokens, - char_to_idx, - } - } - - // --- 业务策略方法 --- - - /// 将字符转为索引，不存在返回 -1 (保持与原 Python 库行为一致) - pub fn char_to_index(&self, char_str: &str) -> i32 { - if let Some(&idx) = self.char_to_idx.get(char_str) { - idx as i32 - } else { - -1 - } - } - - /// 将索引转为字符引用，零拷贝。若越界返回 None - pub fn index_to_char_ref(&self, index: usize) -> Option<&str> { - self.tokens.get(index).map(|cow| cow.as_ref()) - } - - pub fn is_valid_char(&self, char_str: &str) -> bool { - self.char_to_idx.get(char_str).is_some() - } - pub fn size(&self) -> usize { - self.tokens.len() - } -} - -// ========================================== -// 4. 标准 Display 接口实现 (对应 __str__) -// ========================================== -impl std::fmt::Display for Charset { - fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result { - write!(f, "Charset [Total Size: {}", self.size(),) - } -} +// /// 从外部外部 JSON 文件动态加载字符集(在后续优化中移除) +// pub fn from_json_file>(path: P) -> anyhow::Result { +// let path = path.as_ref(); +// if !path.exists() { +// return Err(anyhow!("模型元数据配置文件不存在: {:?}", path)); +// } +// +// let mut file = File::open(path)?; +// let mut content = String::new(); +// file.read_to_string(&mut content)?; +// +// let dto: ModelMetadataDto = serde_json::from_str(&content) +// .map_err(|e| anyhow!("JSON 反序列化失败，请检查字段是否完整: {}", e))?; +// +// // 1. 将 DTO 的字符串数组转化为强类型的 Charset +// let tokens: Vec> = +// dto.charset.into_iter().map(|s| Cow::Owned(s)).collect(); +// let charset = Charset::new(tokens); +// +// // 2. 解析 resize 策略（重现 Python 的复杂条件判断） +// if dto.resize.len() != 2 { +// return Err(anyhow!( +// "'resize (or image)' 字段必须是包含两个元素的数组，例如 [-1, 64]" +// )); +// } +// let r0 = dto.resize[0]; +// let r1 = dto.resize[1]; +// +// let resize = if r0 == -1 { +// if dto.word { +// // 如果 word 为 true，且包含 -1，Python 里是 resize 为 (r1, r1) 的正方形 +// Resize::Square(r1 as u32) +// } else { +// // 如果 word 为 false，且包含 -1，Python 里是高度固定为 r1，宽度按原图比例缩放 +// Resize::DynamicWidth(r1 as u32) +// } +// } else { +// // 正常的固定宽高 +// Resize::Fixed(r0 as u32, r1 as u32) +// }; +// +// Ok(Self { +// charset, +// word: dto.word, +// resize, +// channel: dto.channel, +// normalization: dto.normalization, +// }) +// } \ No newline at end of file diff --git a/tests/ocr_test.rs b/tests/ocr_test.rs index 467c0fc..cee7d53 100644 --- a/tests/ocr_test.rs +++ b/tests/ocr_test.rs @@ -1,8 +1,11 @@ -use ddddocr_rs::{Ocr, Slider, Detector, ModelMetadata, OcrSession, DetBuilder, DetSession}; // 假设你的包名是这个 +use ddddocr_rs::models::det::DetectionResult; +use ddddocr_rs::{DetBuilder, DetSession, Detector, ModelMetadata, Ocr, OcrSession, Slider}; // 假设你的包名是这个 use image::{DynamicImage, Rgb}; use std::fs; use std::path::Path; -use ddddocr_rs::models::det::DetectionResult; +mod char_slice; +use char_slice::CHARSET_BETA; +use ddddocr_rs::models::ocr::metadata::{Normalization, Resize}; fn load_image>(path: P) -> anyhow::Result { // 1. 先将泛型转为具体的 &Path 引用 @@ -16,8 +19,8 @@ fn load_image>(path: P) -> anyhow::Result { } /// 将检测结果绘制在图像上并保存 fn save_debug_image( - dynamic_img: &DynamicImage, // 【优化点 1】直接传入解码好的引用，拒绝重复解码 - bboxes: &[DetectionResult], // 【修改点 1】类型改为自定义结构体切片 + dynamic_img: &DynamicImage, // 【优化点 1】直接传入解码好的引用，拒绝重复解码 + bboxes: &[DetectionResult], // 【修改点 1】类型改为自定义结构体切片 output_path: &str, ) -> anyhow::Result<()> { // 删除了原本的 let dynamic_img = image::load_from_memory(image_bytes)?; @@ -60,17 +63,29 @@ fn save_debug_image( Ok(()) } - #[test] fn test_full_classification() { // 1. 初始化模型 - let ocr = OcrSession::new("D:\\CNWei\\CNW\\Rust\\ddddocr-rs\\models\\common_sml2h3_f32.onnx",ModelMetadata::from_builtin_beta()).expect("模型加载失败"); + let ocr = OcrSession::new( + "D:\\CNWei\\CNW\\Rust\\ddddocr-rs\\models\\common_sml2h3_f32.onnx", + ModelMetadata::from_static_slice( + CHARSET_BETA, + false, + Resize::DynamicWidth(64), + 1, + Normalization::MinusOneToOne, + ), + ) + .expect("模型加载失败"); // 2. 加载测试图片 let img = image::open("samples/code2.png").expect("测试图片不存在"); // 3. 执行识别 - let result = Ocr::new(&ocr).predict(&img).expect("识别过程出错").into_text(); + let result = Ocr::new(&ocr) + .predict(&img) + .expect("识别过程出错") + .into_text(); println!("识别结果: {}", result); assert!(!result.is_empty()); @@ -101,10 +116,7 @@ fn test_det_load() -> anyhow::Result<()> { for (i, bbox) in bboxes.iter().enumerate() { // 【修改点 3】将原来的 bbox[0].. 索引访问改为结构体字段访问 - println!( - "目标 [{}]: x1={}, y1={}, x2={}, y2={}, 分数={:.4}, 类别ID={}", - i, bbox.x1, bbox.y1, bbox.x2, bbox.y2, bbox.score, bbox.class_id - ); + println!("目标 [{}]: {}", i, bbox); } } Ok(()) @@ -129,9 +141,7 @@ fn test_real_slide_match() { // 3. 打印结果 println!("-------------------------------------------"); - println!("滑块匹配测试结果:"); - println!("检测坐标: [x: {}, y: {}]", result.target_x, result.target_y); - println!("置信度: {:.4}", result.confidence); + println!("{}", result); println!("耗时: {:?}", duration); println!("-------------------------------------------");