Identification and Estimation of the Taylor Cone in Electrospinning Technique Using the Yolo Network

Nanoscience and nanotechnology are revolutionizing various scientific and industrial fields by exploring materials at the nanoscale. This study focuses on the electro spinning process of nanofibers, essential in the manufacture of medical textiles, ceramics, and other products. The “Taylor Cone,” a visible phenomenon under high electrical field, is important for evaluating the quality of electrospinning. This article presents a method to geometrically identify and estimate the Taylor Cone using process images captured in the electronics laboratory of the School of Physics at the National University of Trujillo. Experimental parameters included a power source of 0–12 kVolts and an 8 cm distance to the collector plate. The dataset was categorized into “Ausencia,” “Formacion,” and “Cono_Taylor.” Using Yolov8n after 150 epochs, a reliability of 70% in predictions was achieved. Results showed precision rates of 88.24 %, 70.3%, and 76.62%, with recalls of 84.5%, 100%, and 84.5%. Corresponding F1-scores were 86.33%, 82.56%, and 79.68%. The overall accuracy was 77.24%. For geometric estimation, an image from the “Formacion” class revealed a Taylor Cone height of 0.9 mm, a radius of 0.35 mm, and angles measured at <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$23.776^\circ,60.113^\circ$</tex>, and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$96.111^{\circ}$</tex> for the cone vertices. This study highlights the effectiveness of Yolov8n in detecting and characterizing the Taylor Cone, thereby enhancing the precision and efficiency of electrospinning in nanotechnology and biomedicine applications.