Update README.md

README.md

@@ -11,37 +11,38 @@ tags:
 
 # LiteALPR Models
 
-Pretrained weights for **LiteALPR**, an ultra-fast, open-source Automatic License Plate Recognition (ALPR) library explicitly optimized for high-throughput execution.
+Pretrained weights for **LiteALPR**, a lightweight open-source library for fast and accurate Automatic License Plate Recognition (ALPR), explicitly optimized for high-throughput execution on both edge and cloud environments.
 
 ## Model Details
 - **Architecture**:
   - **Detector**: YOLOv8n-Efficient (utilizing Ghost Convolution via lightweight C3Ghost blocks).
-  - **Recognizer**: SVTR26-Tiny (featuring an Efficient RCTC Decoder to bypass heavy attention mechanisms).
+  - **Recognizer**: SVTR26-Tiny (featuring an Efficient RCTC Decoder to bypass computationally heavy attention mechanisms).
 - **License**: GNU Affero General Public License v3.0 (AGPL-3.0)
+- **Repository**: [LiteALPR GitHub](https://github.com/vn-anhnth/LiteALPR)
 
 ## Training Data
-The models were trained and validated on a large-scale aggregated dataset of 52,595 images, encompassing:
-- **Brazil**: RodoSol-ALPR dataset.
-- **China**: CBLPRD dataset.
-- **Vietnam**: Public datasets and self-collected real-world traffic footage.
+To ensure robust, multi-national generalization, the models were trained and validated on a large-scale aggregated dataset of 52,595 images from diverse geographic regions:
+- **Brazil**: [RodoSol-ALPR Dataset](https://github.com/laroca/RodoSol-ALPR)
+- **China**: [CBLPRD Dataset](https://github.com/detectRecog/CBLPRD)
+- **Vietnam**: [Kaggle License Plates Dataset](https://www.kaggle.com/datasets/duydieunguyen/licenseplates) alongside proprietary self-collected real-world traffic footage.
 
-Half of the recognition crops were synthetically degraded (motion blur, Gaussian noise, perspective skew) to ensure high robustness in real-world scenarios.
+Half of the recognition crops were synthetically degraded (e.g., motion blur, Gaussian noise, perspective skew) during training to ensure high resilience against challenging real-world scenarios.
 
 ## Evaluation Results
 - **Detection (YOLOv8n-Efficient)**: 99.45% mAP@50, 5.69 GFLOPs, 2.00M parameters.
 - **Recognition (SVTR26-Tiny)**: 89.15% Sequence Accuracy, 3.28% Character Error Rate (CER), 4.22M parameters.
-
-*Note: Tested on NVIDIA RTX 3060 and AMD Ryzen 5 CPU in ONNX FP32 format.*
+- **End-to-End Speed**: 66.5 FPS on GPU (NVIDIA RTX 3060) and 23.4 FPS on CPU (AMD Ryzen 5 4600G) using ONNX FP32 execution.
 
 ## Known Limitations
-- The model is currently optimized and validated specifically for license plates from **Vietnam, Brazil, and China**. Recognition accuracy may degrade for other regional formats without fine-tuning.
+- The model currently provides highly accurate out-of-the-box support for **single-line alphanumeric license plates** (A–Z, Đ, 0–9) typical of many global regions. 
+- Recognition accuracy on heavily curved multi-line plates or non-Latin characters (e.g., Cyrillic, Arabic) may require custom fine-tuning using the provided training scripts.
 
 ## How to Cite
 If you use these models in your research, please cite our paper:
 
 ```bibtex
 @article{nguyen2026litealpr,
-  title={LiteALPR: An ultra-fast and accurate open-source license plate recognition library},
+  title={LiteALPR: A lightweight open-source library for fast and accurate license plate recognition},
   author={Nguyen, Anh and Hoang, Vinh Truong},
   journal={SoftwareX},
   year={2026}