id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_nvdskcxcwfebbhpl6q2kliwoai	Dan López-Puigdollers	Recognizing white blood cells with local image descriptors	2019	16	.pdf	application/pdf	10255	876	64	(geometric) features, we explore the possibilities of local image descriptors, since they are a simple approachthey require no explicit The results indicate that the approach is encouraging, and that both the sparse keypoint detectors and the dense regular sampling Key words: White blood cells recognition, local image descriptors, SIFT, interest point detectors, visual vocabulary that segmentation itself is a non-trivial problem, and it is generally hard to produce robust and reliable segmentations, being a critical preliminary step for subsequent feature computations (Bikhet, Darwish, Tolba & Shaheen, 2000; Sarrafzadeh, An interesting alternative are local image descriptors computed at interest points, and the use of a pooling strategy for "detection" approaches, a common local descriptor, the wellknown Scale Invariant Feature Transform (SIFT) (Lowe, 2004), Figure 1: Keypoints detected with different detectors on example images of different white blood cell types Yang, Betteridge, Carlson, Mishra, Gardner, Kisiel, Krishnamurthy, Lao, Mazaitis, Mohamed, Nakashole, Platanios, Ritter,	./cache/work_nvdskcxcwfebbhpl6q2kliwoai.pdf	./txt/work_nvdskcxcwfebbhpl6q2kliwoai.txt