Identification of hard exudates in retinal images.

被引:0
|
作者
Dhiravidachelvi, E. [1 ]
Rajamani, V [2 ]
Janakiraman, P. A. [3 ]
机构
[1] Sathyabama Univ, Dept Elect & Commun Engn, Madras, Tamil Nadu, India
[2] Veltech Multi Tech Engn Coll, Dept Elect & Commun Engn, Madras, Tamil Nadu, India
[3] IIT, Madras, Tamil Nadu, India
来源
BIOMEDICAL RESEARCH-INDIA | 2017年 / 28卷
关键词
Hard exudates; Soft exudates; Optic disc; Hue; Parameterization; Retinal image;
D O I
暂无
中图分类号
R318 [生物医学工程];
学科分类号
0831 ;
摘要
A new procedure for identification of the hard exudates is described in this Paper. Images are sorted into as many as 15 types using the maximum value of the hue in an image. Hue (h) at a point is defined as the ratio of green (g) to red (r) intensities. Given an image, the type-number is assigned to it. The exudates are also identified by using the hue, h. The varying intensities have been taken care of by the linear fit obeyed by the plot: h = m*g+c. A modified hue variable `Bh' is used to eliminate the soft exudates which have a blue component, by fitting lines like Bh=w1*h+w2. Very low intensity yellow coloured patches which do not qualify as hard exudates are removed by a discriminating threshold `de'. The parameters like (m, c), ( w1, w2), de are listed for the 15 types, in a Look up Table derived from experiments. The table entries vary in a structured manner. The tables can further be simplified as expressions if desired.
引用
收藏
页码:S336 / S343
页数:8
相关论文
共 50 条
  • [41] FEATURE-BASED REGISTRATION OF RETINAL IMAGES.
    Peli, Eli
    Augliere, Reed A.
    Timberlake, George T.
    IEEE Transactions on Medical Imaging, 1987, MI-6 (03) : 272 - 278
  • [42] The role of domain knowledge in the detection of retinal hard exudates
    Hsu, W
    Pallawala, PMDS
    Lee, ML
    Eong, KGA
    2001 IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, VOL 2, PROCEEDINGS, 2001, : 246 - 251
  • [43] Information Content and Identification of Infrared Images.
    Imre, Benko
    Meres es automatika, 1981, 29 (01): : 2 - 8
  • [44] Bin loss for hard exudates segmentation in fundus images
    Guo, Song
    Wang, Kai
    Kang, Hong
    Liu, Teng
    Gao, Yingqi
    Li, Tao
    NEUROCOMPUTING, 2020, 392 : 314 - 324
  • [45] HARD COPY FOR TRUE THREE-DIMENSIONAL IMAGES.
    Hodges, Larry F.
    Love, Shaun
    McAllister, David F.
    Information Display, 1987, 3 (08) : 12 - 15
  • [46] Improved Sparse Coded Features for Automatic Identification and Discrimination of Exudates and Drusen in Retinal Fundus Images
    Mukesh Kumar
    Kumi Rani
    SN Computer Science, 6 (1)
  • [47] Automated Detection of Optic Disc and Exudates in Retinal Images
    Siddalingaswamy, P. C.
    Prabhu, K. Gopalakrishna
    13TH INTERNATIONAL CONFERENCE ON BIOMEDICAL ENGINEERING, VOLS 1-3, 2009, 23 (1-3): : 277 - +
  • [48] Automatic detection of optic disc and exudates in retinal images
    Kavitha, D
    Devi, SS
    2005 INTERNATIONAL CONFERENCE ON INTELLIGENT SENSING AND INFORMATION PROCESSING, PROCEEDINGS, 2005, : 501 - 506
  • [49] An Investigation into Neural Networks for the Detection of Exudates in Retinal Images
    Schaefer, Gerald
    Leung, Edmond
    APPLICATIONS OF SOFT COMPUTING, 2009, 52 : 169 - 177
  • [50] A new retinal thickness measurement algorithm for OCT images.
    Boyer, KL
    Koozekanani, D
    Roberts, C
    Katz, SE
    INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE, 2001, 42 (04) : S701 - S701