Identification the shape of biconcave Red Blood Cells using Histogram of Oriented Gradients and covariance features

In this paper, a novel methodology for estimating the shape of human biconcave Red Blood Cells (RBCs), using color scattering images, is presented. The information retrieval process includes, image normalization, features extraction using both Histogram of Oriented Gradients (HoG) and region covariance features (RCoV); and features dimensionality reduction using the Independent Component Analysis (ICA). The points of interest (PoIs) are detected using the Harriscorner detector in order to extract the image features. A scheme using adjustable algorithms, i.e. support vectors machine (SVM) is adopted in order to fuse the multimodal features. A Radial Basis Function Neural Network (RBF-NN) estimates the RBC geometrical properties. The proposed method is evaluated in both regression and identification tasks by processing images of a simulated device used to acquire scattering phenomena of moving RBCs. The evaluation database includes 23625 scattering images, obtained by means of the Boundary Element Method. The regression and identification accuracy of the actual RBC shape is estimated using three feature sets in the presence of additive white Gaussian noise from 60 to 10 dB SNR, giving a mean error rate less than 1 percent of the actual RBC shape, and more than 99 percent mean identification rate in a set of valid RBCs size.