Darko Stefanovic

Deep Neural Networks Based Recognition of Plant Diseases by Leaf Image Classification

The latest generation of convolutional neural networks (CNNs) has achieved impressive results in the field of image classification. This paper is concerned with a new approach to the development of plant disease recognition model, based on leaf image classification, by the use of deep convolutional networks. Novel way of training and the methodology used facilitate a quick and easy system implementation in practice. The developed model is able to recognize 13 different types of plant diseases out of healthy leaves, with the ability to distinguish plant leaves from their surroundings. According to our knowledge, this method for plant disease recognition has been proposed for the first time. All essential steps required for implementing this disease recognition model are fully described throughout the paper, starting from gathering images in order to create a database, assessed by agricultural experts. Caffe, a deep learning framework developed by Berkley Vision and Learning Centre, was used to perform the deep CNN training. The experimental results on the developed model achieved precision between 91% and 98%, for separate class tests, on average 96.3%.

Breast volume estimation from systematic series of CT scans using the Cavalieri principle and 3D reconstruction

OBJECTIVES Preoperative breast volume estimation is very important for the success of the breast surgery. In the present study, two different breast volume determination methods, Cavalieri principle and 3D reconstruction were compared. MATERIAL AND METHODS Consecutive sections were taken in slice thickness of 5 mm. Every 2nd breast section in a set of consecutive sections was selected. We marked breast tissue with blue line on each selected section, and so prepared CT scans used for breast volume estimation. The volumes of the 60 breasts were estimated using the Cavalieri principle and 3D reconstruction. RESULTS The mean breast volume value was established to be 467.79 ± 188.90 cm(3) with Cavalieri method and 465.91 ± 191.41 cm(3) with 3D reconstruction. The mean CE for the estimates in this study was calculated as 0.25%. Skin-sparing volume was about 91.64% of the whole breast volume. Both methods are very accurate and have a strong linear association. CONCLUSION Our results suggest that the calculation of breast volume or its part in vivo from systematic series of CT scans using the Cavalieri principle or 3D breast reconstruction is accurate enough to have a significant clinical benefit in planning reconstructive breast surgery. These methods can help the surgeon guide the choice of the most appropriate implant or/and flap preoperatively.

Evaluating the role of content in subjective video quality assessment.

Video quality as perceived by human observers is the ground truth when Video Quality Assessment (VQA) is in question. It is dependent on many variables, one of them being the content of the video that is being evaluated. Despite the evidence that content has an impact on the quality score the sequence receives from human evaluators, currently available VQA databases mostly comprise of sequences which fail to take this into account. In this paper, we aim to identify and analyze differences between human cognitive, affective, and conative responses to a set of videos commonly used for VQA and a set of videos specifically chosen to include video content which might affect the judgment of evaluators when perceived video quality is in question. Our findings indicate that considerable differences exist between the two sets on selected factors, which leads us to conclude that videos starring a different type of content than the currently employed ones might be more appropriate for VQA.

FaceTime — Deep learning based face recognition attendance system

In the interest of recent accomplishments in the development of deep convolutional neural networks (CNNs) for face detection and recognition tasks, a new deep learning based face recognition attendance system is proposed in this paper. The entire process of developing a face recognition model is described in detail. This model is composed of several essential steps developed using todays most advanced techniques: CNN cascade for face detection and CNN for generating face embeddings. The primary goal of this research was the practical employment of these state-of-the-art deep learning approaches for face recognition tasks. Due to the fact that CNNs achieve the best results for larger datasets, which is not the case in production environment, the main challenge was applying these methods on smaller datasets. A new approach for image augmentation for face recognition tasks is proposed. The overall accuracy was 95.02% on a small dataset of the original face images of employees in the real-time environment. The proposed face recognition model could be integrated in another system with or without some minor alternations as a supporting or a main component for monitoring purposes.

Suppression of Metal Artefacts in CT Using Virtual Singorams and Corresponding MR Images

Medical imaging is invaluable when it comes to gaining insight into the human body. As is well known, medical images need to deal with artefacts. This article presents a modern procedure for metal artifact reduction in computed tomography, which relies on additional information extracted from corresponding magnetic resonance images. We conducted a simulation study so as to compare the resulting images with those corrected, using the baseline linear interpolation method. The outcome indicates that the proposed method incomparably outperforms the baseline and reduces metal artefacts, improving the quality of images, which can be later used in a clinical setting.