TY - JOUR AU - Hongthong, Thananath AU - Phoomvuthisarn, Suronapee PY - 2021/12/28 Y2 - 2024/03/29 TI - Flower Image Segmentation using Saliency Map with the Application of HSV Color Space and Color Mask JF - JOURNAL OF INFORMATION SCIENCE AND TECHNOLOGY JA - JIST VL - 11 IS - 2 SE - Research Article: Soft Computing (Detail in Scope of Journal) DO - 10.14456/jist.2021.15 UR - https://ph02.tci-thaijo.org/index.php/JIST/article/view/245602 SP - 38-48 AB - <p>The classification of flowers is a challenging task, due to the similarity of flowers’ physical characteristics. Image segmentation techniques can simplify such details within the image background, making it possible to classify flowers efficiently. In this paper, we purpose a technique for image segmentation based on saliency map to select interested region within the image. The use of saliency map combining with the HSV color space with color mask can reduce insignificant details within the image background. Experimental results have shown that our method can select interested region and can reduce the background detail considerably. Our method can achieve 54% mean IoU (up to 13% higher than previous works), while achieving accuracy, precision, recall and F1 values at 87% when it integrates efficiency with the VGG-16 pre-trained model.</p> ER -