TY - JOUR ID - 139836 TI - Automatic Image Cropping and Semantic Object Selection JO - Journal of Applied Intelligent Systems and Information Sciences JA - JAISIS LA - en SN - 2821-1987 AU - Kakaei, Serveh AD - Department of Industrial Engineering, University of Kurdistan, Sanandaj, Iran Y1 - 2021 PY - 2021 VL - 2 IS - 2 SP - 22 EP - 32 KW - Image Cropping KW - Object Selection KW - Image Segmentation KW - Semantic segmentation DO - 10.22034/jaisis.2021.307161.1035 N2 - Automatic image cropping is an important method for changing the visual quality of digital photos without using to tedious manual selection. Operation of cropping is common to the photographic, graphic design, film developing. At the process of cropping undesired regions cuts away. The performance of an intelligent image cropper highly depends on the ability to detect objects and speed in cutting operation. A lot of methods have been proposed to automate the cropping operation but the most important thing in object cropping is to identify the objects in the image. There are many techniques to detect objects in an image like object detection, image processing and image segmentation. Image segmentation divides a digital image into multiple segments and in this process one label assigns to every pixel in an image. The goal of segmentation is to simplify change the representation of an image into something that is more meaningful and easier to analyze. In this paper, history of development in image segmentation was investigated based on discontinuity and similarity detection-based approaches. We investigated advantage and disadvantage of image segmentation and we studied new models in semantic segmentation and interactive object segmentation. UR - https://journal.research.fanap.com/article_139836.html L1 - https://journal.research.fanap.com/article_139836_742e0c448e65aa0602fc037d385fab71.pdf ER -