MI-CAPTCHA: Enhance the Security of CAPTCHA Using Mooney Images
Authors: Jingmeng Li, Lukang Fu, Surun Yang, Hui Wei
AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We experimentally demonstrate that Hi MI performs better than other baseline methods in terms of both image quality and application potential in two MI-CAPTCHA schemes. Additionally, we conduct experiments to explore the solving performance of humans and CAPTCHA solvers under different parameter settings of schemes, providing valuable reference for the practical application. |
| Researcher Affiliation | Academia | Jingmeng Li, Lukang Fu, Surun Yang, Hui Wei* Laboratory of Algorithms for Cognitive Models, Fudan University, Shanghai, China EMAIL, EMAIL |
| Pseudocode | No | The paper describes the Hi MI framework and its components (CQNet, PDA module) through text and diagrams, but it does not include any explicitly labeled pseudocode blocks or algorithms. |
| Open Source Code | No | The paper does not contain an explicit statement or a direct link to a code repository for the methodology described. |
| Open Datasets | Yes | Animal 2K (Li et al. 2022) is used in image matting and consists of 2,000 highresolution images, with 1,800 images in the training set and 200 images in the test set. MS COCO 2017 (Lin et al. 2014) is a classical dataset for multiple computer vision tasks such as object detection and instance segmentation. |
| Dataset Splits | Yes | Animal 2K (Li et al. 2022) is used in image matting and consists of 2,000 highresolution images, with 1,800 images in the training set and 200 images in the test set. |
| Hardware Specification | Yes | CQNet is implemented in Python with Py Torch and is trained on a Linux server with NVIDIA 3090 GPU. The rests are implemented in MATLAB 2021a on a machine with Intel Core i5-3470 CPU @ 3.20 GHz and 16 GB of main memory. |
| Software Dependencies | Yes | CQNet is implemented in Python with Py Torch and is trained on a Linux server with NVIDIA 3090 GPU. The rests are implemented in MATLAB 2021a on a machine with Intel Core i5-3470 CPU @ 3.20 GHz and 16 GB of main memory. |
| Experiment Setup | Yes | For training stage, we use a batch size of 16 and train the model for 300 epochs with an initial learning rate of 0.01. |