Advancing Bag-of-visual-words Representations For Lesion Classification In Retinal Images.

Diabetic Retinopathy (DR) is a complication of diabetes that can lead to blindness if not readily discovered. Automated screening algorithms have the potential to improve identification of patients who need further medical attention. However, the identification of lesions must be accurate to be useful for clinical application. The bag-of-visual-words (BoVW) algorithm employs a maximum-margin classifier in a flexible framework that is able to detect the most common DR-related lesions such as microaneurysms, cotton-wool spots and hard exudates. BoVW allows to bypass the need for pre- and post-processing of the retinographic images, as well as the need of specific ad hoc techniques for identification of each type of lesion. An extensive evaluation of the BoVW model, using three large retinograph datasets (DR1, DR2 and Messidor) with different resolution and collected by different healthcare personnel, was performed. The results demonstrate that the BoVW classification approach can identify different lesions within an image without having to utilize different algorithms for each lesion reducing processing time and providing a more flexible diagnostic system. Our BoVW scheme is based on sparse low-level feature detection with a Speeded-Up Robust Features (SURF) local descriptor, and mid-level features based on semi-soft coding with max pooling. The best BoVW representation for retinal image classification was an area under the receiver operating characteristic curve (AUC-ROC) of 97.8% (exudates) and 93.5% (red lesions), applying a cross-dataset validation protocol. To assess the accuracy for detecting cases that require referral within one year, the sparse extraction technique associated with semi-soft coding and max pooling obtained an AUC of 94.2 ± 2.0%, outperforming current methods. Those results indicate that, for retinal image classification tasks in clinical practice, BoVW is equal and, in some instances, surpasses results obtained using dense detection (widely believed to be the best choice in many vision problems) for the low-level descriptors.

Mapeamento de semeaduras de soja (Glycine max (L.)Merr.) mediante dados MODIS/Terra E TM/Landsat 5: um comparativo

The objective of this work was to compare the soybean crop mapping in the western of Parana State by MODIS/Terra and TM/Landsat 5 images. Firstly, it was generated a soybean crop mask using six TM images covering the crop season, which was used as a reference. The images were submitted to Parallelepiped and Maximum Likelihood digital classification algorithms, followed by visual inspection. Four MODIS images, covering the vegetative peak, were classified using the Parallelepiped method. The quality assessment of MODIS and TM classification was carried out through an Error Matrix, considering 100 sample points between soybean or not soybean, randomly allocated in each of the eight municipalities within the study area. The results showed that both the Overall Classification (OC) and the Kappa Index (KI) have produced values ranging from 0.55 to 0.80, considered good to very good performances, either in TM or MODIS images. When OC and KI, from both sensors were compared, it wasn't found no statistical difference between them. The soybean mapping, using MODIS, has produced 70% of reliance in terms of users. The main conclusion is that the mapping of soybean by MODIS is feasible, with the advantage to have better temporal resolution than Landsat, and to be available on the internet, free of charge.