{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,20]],"date-time":"2026-03-20T10:34:06Z","timestamp":1774002846720,"version":"3.50.1"},"reference-count":35,"publisher":"PeerJ","license":[{"start":{"date-parts":[[2026,3,20]],"date-time":"2026-03-20T00:00:00Z","timestamp":1773964800000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>This research presents a portable, edge-optimized system designed to overcome the limitations of traditional surveillance in detecting modern threats such as disguised individuals, hidden weapons, and flying drones, particularly in environments with limited resources. The system leverages a modified Florence v2 Vision-Language Transformer for military-grade object detection, with its core deployed on a Raspberry Pi 5. A separate ESP32-S2 DevKit handles wireless alerts using the ESP-NOW protocol for fast, local communication. To achieve high accuracy with low computational overhead, we adapted the Florence v2\u00a0model using Parameter Efficient Fine-Tuning (PEFT) and Low-Rank Adaptation (LoRA). The model was trained on a dataset of 12,000 images, augmented with manual annotations and synthetic images generated by Generative Adversarial Networks (GANs). The final model achieved an overall accuracy of 88.01%, outperforming other popular models such as You Only Look Once version 8 (YOLOv8) and Fast Region-based Convolutional Neural Network (FRCNN). Our model achieved a Mean Average Precision (mAP@50) score of 64.83, compared to YOLOv8 (64) and FRCNN (56). After training, the model was optimized through quantization and pruning to ensure efficient execution on edge hardware. It processes a frame in 16.6 s, utilizing only 38.4% of the Raspberry Pi\u2019s RAM and 50.9% of its CPU. The integrated ESP32-S2\u00a0module enables the system to transmit critical alerts over long distances without the need for cloud connectivity or a central server. This system demonstrates the capability of autonomous, real-time surveillance across critical domains including military borders, urban airspaces, and civilian infrastructure.<\/jats:p>","DOI":"10.7717\/peerj-cs.3579","type":"journal-article","created":{"date-parts":[[2026,3,20]],"date-time":"2026-03-20T08:07:51Z","timestamp":1773994071000},"page":"e3579","source":"Crossref","is-referenced-by-count":0,"title":["Edge-optimized threat detection with Florence v2 for autonomous surveillance in resource-constrained environments"],"prefix":"10.7717","volume":"12","author":[{"given":"Cynthia","family":"Konar","sequence":"first","affiliation":[]},{"given":"Diya","family":"Das","sequence":"additional","affiliation":[]},{"given":"Subitha","family":"D.","sequence":"additional","affiliation":[]},{"given":"Kavitha","family":"Jc","sequence":"additional","affiliation":[]},{"given":"Sweety","family":"Singh","sequence":"additional","affiliation":[]}],"member":"4443","published-online":{"date-parts":[[2026,3,20]]},"reference":[{"key":"10.7717\/peerj-cs.3579\/ref-1","doi-asserted-by":"publisher","first-page":"86","DOI":"10.1109\/RTEICT.2017.8256563","article-title":"Implementation of spy robot for a surveillance system using internet protocol of raspberry pi","author":"Abdalla","year":"2017"},{"key":"10.7717\/peerj-cs.3579\/ref-2","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/ICTC55196.2022.9952832","article-title":"Peer to peer communication for the internet of things using ESP32\u00a0microcontroller for indoor environments","author":"Abdul","year":"2022"},{"key":"10.7717\/peerj-cs.3579\/ref-3","doi-asserted-by":"publisher","DOI":"10.13140\/RG.2.2.23650.54722","article-title":"Violence detection using vision transformer and explainable AI","author":"Abdullah","year":"2024","journal-title":"Preprint"},{"issue":"13","key":"10.7717\/peerj-cs.3579\/ref-4","doi-asserted-by":"publisher","first-page":"5849","DOI":"10.3390\/s23135849","article-title":"State-of-the-art deep learning methods for object detection in remote sensing satellite images","volume":"23","author":"Adegun","year":"2023","journal-title":"Sensors"},{"issue":"3","key":"10.7717\/peerj-cs.3579\/ref-5","doi-asserted-by":"publisher","first-page":"458","DOI":"10.3390\/rs12030458","article-title":"Comparative research on deep learning approaches for airplane detection from very high-resolution satellite images","volume":"12","author":"Alganci","year":"2020","journal-title":"Remote Sensing"},{"key":"10.7717\/peerj-cs.3579\/ref-6","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2501.13066","article-title":"SMART-Vision: survey of modern action recognition techniques in vision","author":"AlShami","year":"2025"},{"key":"10.7717\/peerj-cs.3579\/ref-7","doi-asserted-by":"publisher","first-page":"104630","DOI":"10.1016\/j.micpro.2022.104630","article-title":"Aircraft detection in satellite imagery using deep learning-based object detectors","volume":"94","author":"Azam","year":"2022","journal-title":"Microprocessors and Microsystems"},{"key":"10.7717\/peerj-cs.3579\/ref-8","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/CarpathianCC.2019.8765944","article-title":"Using the ESP32\u00a0microcontroller for data processing","author":"Babiuch","year":"2019"},{"key":"10.7717\/peerj-cs.3579\/ref-9","doi-asserted-by":"publisher","first-page":"126","DOI":"10.1109\/INDICON59947.2023.10440938","article-title":"Object detection for military surveillance using yolo framework","author":"Borthakur","year":"2023"},{"key":"10.7717\/peerj-cs.3579\/ref-10","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/EEITE61750.2024.10654433","article-title":"Illicit object detection in x-ray images using vision transformers","author":"Cani","year":"2024"},{"key":"10.7717\/peerj-cs.3579\/ref-11","doi-asserted-by":"publisher","first-page":"17753","DOI":"10.1109\/JSTARS.2024.3462745","article-title":"Det-YOLO: an innovative high-performance model for detecting military aircraft in remote sensing images","volume":"17","author":"Chen","year":"2024","journal-title":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing"},{"issue":"9","key":"10.7717\/peerj-cs.3579\/ref-12","doi-asserted-by":"publisher","first-page":"4552","DOI":"10.1109\/TCSVT.2023.3281557","article-title":"Cross-drone transformer network for robust single object tracking","volume":"33","author":"Chen","year":"2023","journal-title":"IEEE Transactions on Circuits and Systems for Video Technology"},{"issue":"6","key":"10.7717\/peerj-cs.3579\/ref-13","doi-asserted-by":"publisher","first-page":"3026","DOI":"10.12928\/TELKOMNIKA.v18i6.16174","article-title":"The detection of handguns from live video in real-time based on deep learning","volume":"18","author":"Ghazal","year":"2020","journal-title":"TELKOMNIKA (Telecommunication Computing Electronics and Control)"},{"key":"10.7717\/peerj-cs.3579\/ref-14","doi-asserted-by":"publisher","first-page":"18373","DOI":"10.1109\/CVPR52733.2024.01739","article-title":"End-to-end spatio-temporal action localisation with video transformers","author":"Gritsenko","year":"2024"},{"issue":"11","key":"10.7717\/peerj-cs.3579\/ref-15","doi-asserted-by":"crossref","first-page":"139","DOI":"10.1145\/3422622","article-title":"Generative adversarial networks","volume":"63","author":"Goodfellow","year":"2020","journal-title":"Communications of the ACM"},{"key":"10.7717\/peerj-cs.3579\/ref-16","doi-asserted-by":"publisher","first-page":"216","DOI":"10.37917\/ijeee.21.2.21","article-title":"A comprehensive review for aircraft detection techniques utilizing deep learning","volume":"21","author":"Hameed","year":"2025","journal-title":"Iraqi Journal for Electrical and Electronic Engineering"},{"key":"10.7717\/peerj-cs.3579\/ref-17","doi-asserted-by":"publisher","first-page":"6739","DOI":"10.3390\/s23156739","article-title":"Design and implementation of ESP32-based IoT devices","volume":"23","author":"Hercog","year":"2023","journal-title":"Sensors"},{"key":"10.7717\/peerj-cs.3579\/ref-18","doi-asserted-by":"publisher","first-page":"119223\u2013119237","DOI":"10.1109\/ACCESS.2024.3448619","article-title":"YOLOv8-FDF: a small target detection algorithm in complex scenes","volume":"12","author":"Jiang","year":"2024","journal-title":"IEEE Access"},{"key":"10.7717\/peerj-cs.3579\/ref-19","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/ICRITO61523.2024.10522248","article-title":"AIR-SCAN: aircraft identification and recognition using deep learning scanning","author":"Kumar","year":"2024"},{"issue":"3","key":"10.7717\/peerj-cs.3579\/ref-20","doi-asserted-by":"publisher","first-page":"72","DOI":"10.3390\/jimaging11030072","article-title":"Concealed weapon detection using thermal cameras","volume":"11","author":"Mu\u00f1oz","year":"2025","journal-title":"Journal of Imaging"},{"key":"10.7717\/peerj-cs.3579\/ref-21","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/ISNCC55209.2022.9851713","article-title":"Smart border surveillance system based on deep learning methods","author":"Nakkach","year":"2022"},{"key":"10.7717\/peerj-cs.3579\/ref-22","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/ICCIT60459.2023.10441013","article-title":"Digital border surveillance system: towards illegal migration and trafficking free borders","author":"Noor","year":"2023"},{"key":"10.7717\/peerj-cs.3579\/ref-23","doi-asserted-by":"publisher","first-page":"66","DOI":"10.1016\/j.neucom.2017.09.053","article-title":"Automatic handgun detection alarm in videos using deep learning","volume":"275","author":"Olmos","year":"2018","journal-title":"Neurocomputing"},{"key":"10.7717\/peerj-cs.3579\/ref-24","doi-asserted-by":"publisher","first-page":"24936","DOI":"10.1109\/ACCESS.2022.3149297","article-title":"Lightweight mask R-CNN for warship detection and segmentation","volume":"10","author":"Park","year":"2022","journal-title":"IEEE Access"},{"key":"10.7717\/peerj-cs.3579\/ref-25","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/AIPR.2017.8457969","article-title":"Satellite image classification with deep learning","author":"Pritt","year":"2017"},{"issue":"13","key":"10.7717\/peerj-cs.3579\/ref-26","doi-asserted-by":"publisher","first-page":"6085","DOI":"10.3390\/app11136085","article-title":"Automatic handgun detection with deep learning in video surveillance images","volume":"11","author":"Salido","year":"2021","journal-title":"Applied Sciences"},{"key":"10.7717\/peerj-cs.3579\/ref-27","doi-asserted-by":"publisher","first-page":"54","DOI":"10.1186\/s43067-023-00123-z","article-title":"Object detection using convolutional neural networks and transformer-based models: a review","volume":"10","author":"Shah","year":"2023","journal-title":"Journal of Electrical Systems and Information Technology"},{"issue":"11","key":"10.7717\/peerj-cs.3579\/ref-28","doi-asserted-by":"publisher","first-page":"5830","DOI":"10.3390\/app15115830","article-title":"DeepGun: deep feature-driven one-class classifier for firearm detection using visual gun features and human body pose estimation","volume":"15","author":"Singh","year":"2025","journal-title":"Applied Sciences"},{"key":"10.7717\/peerj-cs.3579\/ref-29","doi-asserted-by":"publisher","first-page":"2805","DOI":"10.1109\/JSTARS.2022.3159981","article-title":"CGC-NET: aircraft detection in remote sensing images based on lightweight convolutional neural network","volume":"15","author":"Wang","year":"2022","journal-title":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing"},{"key":"10.7717\/peerj-cs.3579\/ref-30","doi-asserted-by":"publisher","first-page":"245","DOI":"10.1007\/s00530-024-01423-8","article-title":"Propagating prior information with transformer for robust visual object tracking","volume":"30","author":"Wu","year":"2024","journal-title":"Multimedia Systems"},{"key":"10.7717\/peerj-cs.3579\/ref-31","doi-asserted-by":"publisher","first-page":"4210","DOI":"10.1109\/ICIP.2015.7351599","article-title":"Fast aircraft detection in satellite images based on convolutional neural networks","author":"Wu","year":"2015"},{"key":"10.7717\/peerj-cs.3579\/ref-32","doi-asserted-by":"publisher","first-page":"1156","DOI":"10.1109\/IAEAC50856.2021.9390673","article-title":"Real-time detection of aircraft objects in remote sensing images based on improved yolov4","author":"Yang","year":"2021"},{"key":"10.7717\/peerj-cs.3579\/ref-33","doi-asserted-by":"publisher","first-page":"623","DOI":"10.1109\/ICACCS48705.2020.9074171","article-title":"Border surveillance system using computer vision","author":"Yasar","year":"2020"},{"issue":"9","key":"10.7717\/peerj-cs.3579\/ref-34","doi-asserted-by":"publisher","first-page":"5553","DOI":"10.1109\/TGRS.2016.2569141","article-title":"Weakly supervised learning based on coupled convolutional neural networks for aircraft detection","volume":"54","author":"Zhang","year":"2016","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"key":"10.7717\/peerj-cs.3579\/ref-35","doi-asserted-by":"publisher","first-page":"89","DOI":"10.1109\/ICBASE59196.2023.10303177","article-title":"Research on target detection technology of aircraft satellite images based on improved yolov5\u00a0model","author":"Zhao","year":"2023"}],"container-title":["PeerJ Computer Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/peerj.com\/articles\/cs-3579.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3579.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3579.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3579.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,20]],"date-time":"2026-03-20T08:07:57Z","timestamp":1773994077000},"score":1,"resource":{"primary":{"URL":"https:\/\/peerj.com\/articles\/cs-3579"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,20]]},"references-count":35,"alternative-id":["10.7717\/peerj-cs.3579"],"URL":"https:\/\/doi.org\/10.7717\/peerj-cs.3579","archive":["CLOCKSS","LOCKSS","Portico"],"relation":{},"ISSN":["2376-5992"],"issn-type":[{"value":"2376-5992","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,20]]},"article-number":"e3579"}}