Strabismus, characterized by ocular misalignment and loss of binocular fixation, can result in impaired eye movement, blurred vision, asthenopia, abnormal head posture, reduced stereopsis, and, if untreated in early childhood, amblyopia. Conventional diagnosis depends on in-person ophthalmologic examination and specialized instruments, which limits early detection in resource-constrained settings. In this study, we present an image-based vision-sensing system that transforms a commodity camera into a low-cost ophthalmic screening sensor for automated strabismus detection. Facial images captured by the RGB camera are processed using You Only Look Once v8 Nano (YOLOv8n) deep learning model to localize ocular regions and analyze ocular alignment patterns that reflect the presence of strabismus. The proposed online platform provides a simple web-based user interface and achieves a diagnostic accuracy of 95.48%, demonstrating that consumer-grade imaging sensors, when combined with advanced deep learning algorithms, can function as effective medical screening devices. The results of this work contribute to the advancement of sensor applications in ophthalmology by enabling scalable, noncontact, and remote vision screening, thereby broadening the practical use of imaging sensors and sensor-enabled materials in telemedicine and community-based eye health programs.