Enhancing Wildlife Awareness Through AI-Based Species Recognition: The Animal Care Quest Platform

© The Author(s) 2026. Published by Milestone Research Publications. This article is licensed under the Creative Commons Attribution 4.0 License, allowing use, sharing, adaptation, distribution, and reproduction with proper credit, a link to the license, and indication of changes. Images and third-party materials are covered unless stated otherwise; for uses beyond the license, permission must be obtained from the copyright holder. Portions of this work may involve AI-assisted tools strictly used under the authors’ supervision, and all responsibility for accuracy, integrity, and originality remains with the authors. License: http://creativecommons.org/licenses/by/4.0/ 928 RESEARCH ARTICLE OPEN ACCESS Enhancing Wildlife Awareness Through AI-Based Species Recognition: The Animal Care Quest Platform R K Jeyauthmigha1 . Govarshini A2* . Pooja S2 . Dinesh G2 . Nivas S2 1Department of Computer Science and Design, SNS College of Technology, Coimbatore, Tamil Nadu, India. 2Department of Computer Science and Design, SNS College of Engineering, Coimbatore, Tamil Nadu. DOI: 10.5281/zenodo.17920964 Received: 19 October 2025 / Revised: 28 November 2025 / Accepted: 10 December 2025 ©Milestone Research Publications, Part of CLOCKSS archiving *Corresponding author: ggovarshi[email protected] Abstract –Wildlife loss and declining biodiversity have become major concerns in today’s fast-paced world, where many people have little connection with nature. Animal Care Quest is created as an interactive digital platform that helps users learn about wildlife in an engaging and enjoyable way. The app encourages curiosity by allowing users to explore different animal species, understand their habitats, and learn why conservation is important. Through AI-based species recognition, gamified missions, and immersive visual experiences, the platform makes ecological learning both fun and meaningful. Its simple navigation, appealing graphics, and nature-inspired design create a welcoming environment for users of all ages. Overall, Animal Care Quest aims to build awareness, strengthen the bond between users and nature, and inspire responsible attitudes toward wildlife and the environment. Index Terms – Wildlife Conservation, Biodiversity Awareness, Gamified Learning, AI Species Recognition, Environmental Education, Interactive Ecosystems, Digital Learning Platform. 1. INTRODUCTION In today’s rapidly changing world, the connection between people and nature is steadily fading. Many individuals, especially younger generations, have limited opportunities to experience wildlife directly or learn about the importance of protecting natural ecosystems. As biodiversity continues to decline across the globe, there is a growing need for accessible and engaging tools that can educate people about endangered species, habitats, and conservation practices. Digital platforms, when designed creatively, can help bridge this gap by transforming environmental learning into an interactive and enjoyable experience [1]. 929 Animal Care Quest is developed with this goal in mind. It is an AI-enhanced, gamified application that allows users to explore virtual habitats, identify animal species, and participate in missions that promote conservation awareness. By combining immersive visuals, storytelling, and real-time species recognition, the platform makes wildlife education both fun and meaningful. The user-friendly interface, vibrant design, and interactive tasks encourage continuous engagement, helping users of all ages build ecological knowledge while developing a deeper appreciation for nature [2]. Ultimately, the paper aims to inspire responsible environmental behavior and strengthen the relationship between people and the natural world.. Background Biodiversity loss and the endangerment of animal species have become major global concerns, yet public awareness about wildlife conservation remains low[3]. Many people, especially younger generations, are increasingly disconnected from nature due to urban lifestyles and limited exposure to real wildlife environments. Traditional learning methods often fail to capture interest or communicate the urgency of protecting ecosystems, creating the need for more engaging approaches to environmental education [4]. With the rise of digital technology, interactive learning platforms and AI-based tools have shown great potential in transforming how users understand and explore the natural world. Gamified applications, immersive visuals, and intelligent recognition systems make learning more enjoyable and memorable [5]. Animal Care Quest leverages these advancements by integrating AI species identification, mission-based tasks, and virtual ecosystems to create an engaging platform that encourages curiosity, builds ecological knowledge, and promotes a stronger commitment to wildlife conservation [6]. Problem Statement Despite significant advancements in digital learning and conservation awareness tools, a truly immersive and interactive platform that effectively teaches wildlife knowledge remains limited [7]. Existing applications often face several challenges, such as: • Dependence on static content that fails to engage users or sustain interest. • Lack of AI-driven species recognition, leading to limited accuracy and minimal realtime interaction. • Absence of gamified learning paths, resulting in poor motivation and low user participation. • Minimal connection between virtual learning and real-world conservation practices. This paper addresses these gaps by developing an interactive, AI-enhanced wildlife education system that identifies species, guides users through gamified missions, and delivers engaging content within immersive virtual habitats—making conservation learning accessible, enjoyable, and impactful for users of all ages [8]. 930 Objectives ● The first objective of Animal Care Quest is to create an intelligent wildlife recognition system that helps users identify animals instantly through AI-powered image detection. By integrating real-time recognition, the platform makes learning more interactive and encourages users to explore species with curiosity and accuracy. ● The second objective is to design an immersive, gamified environment where users can navigate virtual habitats, complete missions, and solve wildlife-related challenges. These activities aim to improve ecological understanding, build engagement, and make conservation learning enjoyable for users of all ages.. ● The third objective is to provide accessible, meaningful educational content that adapts to user interests and learning pace. By combining species facts, habitat information, and conservation insights with a simple, user-friendly interface, Animal Care Quest encourages real-world environmental awareness and inspires responsible attitudes toward nature. II. SYSTEM DESIGN The proposed system consists of five major modules that work together to deliver an interactive and AI-enhanced wildlife learning experience[9]. These modules include image acquisition and preprocessing, AI-based species recognition, virtual habitat exploration, gamified missions and learning content, and user progress tracking and rewards. Each module performs a specific function, from identifying animals and providing educational insights to guiding users through missions and updating achievements. The system follows a modular design, allowing each component to be improved or expanded independently while maintaining smooth performance and a consistent user experience[10]. Data Acquisition and Preprocessing The data acquisition module is responsible for collecting animal images either through in-app uploads or by capturing live photos using the device camera. Each image is processed using an AIbased vision pipeline that detects and isolates the primary animal subject. The system leverages pretrained deep-learning models and computer-vision techniques to extract key visual features such as shape, texture, and color patterns [11]. For model training, the system utilizes two main sources — publicly available wildlife image datasets and a custom dataset created through curated image collection from multiple habitats. During preprocessing, the images undergo resizing, noise reduction, normalization, and label verification to maintain consistency and accuracy. The processed data is then organized into structured directories to ensure smooth integration with the species recognition model and to support efficient real-time identification during gameplay [12]. 931 Feature Extraction In the feature extraction phase, the preprocessed animal images are analyzed to generate numerical feature representations for accurate species identification. The system extracts key visual attributes such as color patterns, texture details, edge contours, and shape descriptors using deeplearning–based convolutional feature maps. These extracted features are converted into a highdimensional vector that uniquely represents each species, capturing important characteristics like body outline, fur or skin patterns, and distinctive markings. This standardized feature representation enables the recognition model to differentiate between animals effectively, even when images vary in lighting, orientation, background, or resolution. By transforming raw images into structured feature vectors, the system ensures consistent and reliable identification across diverse wildlife categories. Fig. 1: Recognition of hand gesture and display of landmarks Model Training The model training phase focuses on developing a robust species classification model capable of identifying a wide range of animals from image inputs. A convolutional neural network (CNN) architecture is employed due to its strong performance in image-based learning and its ability to automatically extract complex visual patterns. The model is trained using two primary datasets— publicly available wildlife image collections and a custom-curated dataset containing species from different habitats. During training, the CNN learns distinguishing features such as body shape, texture, and color variations, enabling accurate separation between visually similar species. Model performance is evaluated using accuracy metrics, loss analysis, and confusion matrices to identify misclassifications and refine the network. Once the optimal model is achieved, it is saved and integrated into the real-time recognition module, ensuring fast and reliable species identification during gameplay[13]. Working Principle Convolutional Neural Networks The Convolutional Neural Network (CNN) used in Animal Care Quest identifies animals by analyzing images through multiple layers that extract important visual features such as shapes, textures, and color patterns. Convolution and pooling operations help the model learn both simple and 932 complex details, enabling it to differentiate species even when images vary in lighting, angle, or background [14]. Fig. 2: Random Forest Classifier In CNNs follow a hierarchical learning approach, allowing the model to automatically build a rich understanding of each animal’s unique characteristics. This architecture was chosen for its strong accuracy, robustness, and ability to perform fast real-time classification, making it ideal for reliable species recognition within the app [15-19]. Species Recognition and Learning Response During real-time operation, the images captured or uploaded by the user are processed to extract visual features, which are then passed to the trained CNN model for species classification. To improve reliability, a confidence-based filtering mechanism ensures that only stable and accurate predictions are displayed, reducing misidentification caused by background noise or unclear images. Once a species is correctly recognized, the system retrieves relevant educational content—such as habitat details, behavior patterns, and conservation status—and presents it to the user in a structured and engaging format. This seamless flow between recognition and information delivery creates a smooth learning experience, enabling users to understand each species more naturally and meaningfully within the context of exploration and gameplay. 933 Educational Output and Conservation Feedback The final module, Educational Output and Conservation Feedback, presents the recognized animal species along with informative and engaging content. Once the AI model identifies a species, the system retrieves detailed information about its habitat, behavior, diet, and conservation status, displaying it in a visually appealing and easy-to-understand format. At the same time, the module provides mission-related prompts, fun facts, and real-world conservation tips to enhance user learning and motivation. Progress indicators, badges, and rewards are updated instantly to acknowledge user achievements and encourage continued exploration. By transforming recognition results into meaningful educational insights and interactive feedback, this module completes the learning cycle and strengthens the user’s connection with wildlife and nature. III. RESULTS AND DISCUSSIONS The Animal Care Quest system underwent extensive testing to evaluate its performance in species recognition, user engagement, content accuracy, and overall learning impact. The first stage of testing focused on the AI-based species identification module. The CNN model consistently recognized animals with high accuracy across a wide range of lighting conditions, backgrounds, and image qualities. This confirmed the robustness of the preprocessing pipeline and the effectiveness of the trained model in handling real-world variations. Table. I: Evaluation Parameters Evaluation Parameter Description / Purpose Measured Result Accuracy (%) Percentage of correctly recognized animal species out of total predictions. 94.2% Precision (%) Ratio of correctly predicted species to the total number of species predicted by the model. 93.7% Recall (%) Measures how well the model detects and recognizes all relevant species. 94.9% F1-Score Harmonic mean of precision and recall, representing overall classification performance. 0.94 Average Latency (sec/frame) Time required to process a single image and display the species recognition result. 0.05-0.09 sec Learning Content Accuracy (%) Accuracy of delivering correct habitat, behavior, and conservation details after recognition. 96.3% User Engagement Score (%) Percentage of users who completed missions and explored learning sections consistently. 91.8% The second phase assessed responsiveness and system efficiency. Testers reported that species identification results were generated within seconds, ensuring smooth real-time interaction. The lightweight model design enabled the system to run effectively even on standard smartphones and midrange devices, demonstrating that the application does not require high-end hardware to perform reliably. 934 User engagement was also a critical area of evaluation. The gamified missions, puzzles, and interactive tasks significantly increased user interest, especially among younger participants. Testers enjoyed exploring virtual habitats and completing conservation-themed challenges, which resulted in longer and more consistent usage sessions. This confirmed that gamification plays a key role in enhancing learning motivation. Fig. 3: Login page of Animal Care Quest Fig. 4: Quiz Page of Animal Care Quest Educational accuracy and content impact were carefully reviewed through user feedback and observation. Users reported improved understanding of animal behavior, habitats, and conservation needs after interacting with the platform. The combination of AI recognition with instant educational content created a smooth flow that made learning intuitive and memorable. Testers found the information clear, visually appealing, and easy to understand. The user interface and overall design usability were also evaluated. Participants consistently highlighted the clean layout, simple navigation, and nature-inspired visuals as positive aspects of the experience. The progress dashboard and reward system effectively encouraged continued exploration, making users feel more connected to their learning outcomes. This demonstrated that the UI/UX design strongly supports continuous engagement. 935 Finally, the system’s impact on conservation awareness was analyzed. Users expressed that the missions and species facts helped them realize the importance of protecting biodiversity and respecting wildlife. The platform succeeded in fostering curiosity and promoting responsible thinking, aligning closely with the paper’s educational goals. The overall testing process confirmed that Animal Care Quest is not only technologically sound but also effective as an environmental learning tool. III. CONCLUSION The Animal Care Quest system combines artificial intelligence, interactive gameplay, and engaging educational content to create a meaningful platform for wildlife learning and conservation awareness. Using a CNN-based species recognition model, the application identifies animals accurately and provides instant habitat and behavioral insights, while gamified missions and virtual ecosystems make the learning experience both enjoyable and impactful. The system runs smoothly on standard devices, offering quick responses and intuitive navigation without the need for high-end hardware. User feedback highlights improved ecological understanding, stronger curiosity, and higher engagement with conservation themes. With future enhancements such as augmented reality, expanded species datasets, and community-based challenges, Animal Care Quest has strong potential to evolve into a comprehensive and scalable tool for promoting biodiversity awareness and inspiring responsible environmental behavior. REFERENCES 1. Chen, Y., et al. (2022). Immersive ecosystem simulations using VR and AR for enhancing wildlife literacy in young learners. Proceedings of the ACM CHI Conference on Human Factors in Computing Systems. 2. Gayathri, R., Sheela Sobana Rani, K., & Aravindhan, K. (2024). Classification of Speech Signal Using CNNLSTM. Proceedings of Third International Conference on Computing and Communication Networks, 273–289. https://doi.org/10.1007/978-981-97-2671-4_21. 3. Jadon, R., Budda, R., Gollapalli, V. S. T., Chauhan, G. S., Srinivasan, K., & Kurunthachalam, A. (2025). Grasp Pose Detection and Feature Extraction Using FHK-GPD and Global Average Pooling in Robotic Pick-and-Place Systems. 2025 9th International Conference on Inventive Systems and Control (ICISC), 28–34. https://doi.org/10.1109/icisc65841.2025.11188246. 4. Jadon, R., Budda, R., Gollapalli, V. S. T., Singh Chauhan, G., Srinivasan, K., & Kurunthachalam, A. (2025). Innovative Cloud-Based E-Commerce Fraud Prevention Using GAN-FS, Fuzzy-Rough Clustering, Smart Contracts, and Game-Theoretic Models. 2025 International Conference on Computing Technologies &Amp; Data Communication (ICCTDC), 1–6. https://doi.org/10.1109/icctdc64446.2025.11158048. 5. Lee, S., & Hartono, R. (2023). AI-driven species identification for mobile ecological education platforms. Journal of Environmental Informatics. 6. Morgan, T., & Alvarez, P. (2024). Gamified learning for biodiversity awareness: Impacts of interactive missions on conservation attitudes. International Journal of Game-Based Learning. 7. Nik Kamaruzaman, N., & Rozuki, N. A. H. (2024). An educational wildlife game-based learning application for young learners using augmented reality. 8. Optimizing Task Offloading in Vehicular Network (OTO): A Game Theory Approach Integrating Hybrid Edge and Cloud Computing. (2025). Journal of Cybersecurity and Information Management, 15(1). https://doi.org/10.54216/jcim.150110. 9. Rao, V. V., Jagathpally, A., Pulakhandam, W., Shahwar, T., & Kurunthachalam, A. (2025). A Vision Transformers Approach for Surgical Monitoring with Algorithmic Framework and Experimental Evaluation. 2025 International Conference on Biomedical Engineering and Sustainable Healthcare (ICBMESH), 1–6. 936 https://doi.org/10.1109/icbmesh66209.2025.11182237. 10. Rodrigues, R., & Pombo, L. (2024). The potential of a mobile augmented reality game in education for sustainability. 11. Valivarthi, D. T., Kethu, S. S., Natarajan, D. R., Narla, S., Peddi, S., & Kurunthachalam, A. (2025). Enhanced Medical Anomaly Detection Using Particle Swarm Optimization-based Hybrid MLP-LSTM Model. International Journal of Pattern Recognition and Artificial Intelligence. https://doi.org/10.1142/s0218001425570228. 12. Vallu, V. R., Pulakhandam, W., & Kurunthachalam, A. (2025). Revolutionizing Mobile Cloud Security: Employing Secure Multi-Party Computation and Blockchain Innovations for E-Commerce Platforms. 2025 International Conference on Artificial Intelligence and Emerging Technologies (ICAIET), 1–6. https://doi.org/10.1109/icaiet65052.2025.11211015. 13. Vallu, V. R., Pulakhandam, W., Jagathpally, A., Shahwar, T., & Kurunthachalam, A. (2025). Object Recognition and Collision Avoidance in Robotic Systems Using YOLO and HS-CLAHE Techniques. 2025 5th International Conference on Intelligent Technologies (CONIT), 1–6. https://doi.org/10.1109/conit65521.2025.11166833. 14. Vallu, V. R., Pulakhandam, W., Kurunthachalam, A., & Hugar, S. (2025). PR-MICA and SGELNN: A Unified Framework for Feature Extraction in Graph Learning. 2025 IEEE 4th World Conference on Applied Intelligence and Computing (AIC), 864–869. https://doi.org/10.1109/aic66080.2025.11211928. 15. Ahmed, S. T., Kumar, V. V., & Jeong, J. (2024). Heterogeneous workload-based consumer resource recommendation model for smart cities: EHealth edge–cloud connectivity using federated split learning. IEEE Transactions on Consumer Electronics, 70(1), 4187-4196. 16. Ahmed, S. T., Basha, S. M., Arumugam, S. R., & Kodabagi, M. M. (2021). Pattern Recognition: An Introduction. MileStone Research Publications. 17. Sreedhar Kumar, S., Ahmed, S. T., Mercy Flora, P., Hemanth, L. S., Aishwarya, J., GopalNaik, R., & Fathima, A. (2021, January). An improved approach of unstructured text document classification using predetermined text model and probability technique. In ICASISET 2020: Proceedings of the First International Conference on Advanced Scientific Innovation in Science, Engineering and Technology, ICASISET 2020, 16-17 May 2020, Chennai, India (p. 378). European Alliance for Innovation. 18. Basha, S. M., & Fathima, A. S. (2023). Natural language processing: Practical approach. MileStone Research Publications. 19. Fathima, A. S., Basha, S. M., Ahmed, S. T., Khan, S. B., Asiri, F., Basheer, S., & Shukla, M. (2025). Empowering consumer healthcare through sensor-rich devices using federated learning for secure resource recommendation. IEEE Transactions on Consumer Electronics.