Deep learning approaches to biomedical image segmentation

Learning discriminative features for Facial Expression Recognition (FER) in the wild using Convolutional Neural Networks (CNNs) is a non-trivial task due to the significant intra-class variations and inter-class similarities. Deep Metric Learning (DML) approaches such as center loss and its variants jointly optimized with softmax loss have been adopted in many FER methods to enhance the discriminative power of learned features in the embedding space. However, equally supervising all features with the metric learning method might include irrelevant features and ultimately degrade the generalization ability of the learning algorithm. We propose a Deep Attentive Center Loss (DACL) method to adaptively select a subset of significant feature elements for enhanced discrimination. The proposed DACL integrates an attention mechanism to estimate attention weights correlated with feature importance using the intermediate spatial feature maps in CNN as context. The estimated weights accommodate the sparse formulation of center loss to selectively achieve intra-class compactness and inter-class separation for the relevant information in the embedding space. An extensive study on two widely used wild FER datasets demonstrates the superiority of the proposed DACL method compared to state-of-the-art methods.

/pdf/facial-expression-recognition-in-the-wild-via-deep-attentive-5iuf5i3swg.pdf

Facial Expression Recognition in the Wild via Deep Attentive Center Loss

Crop diseases are a major threat to food security. Identifying the diseases rapidly is still a difficult task in many parts of the world due to the lack of the necessary infrastructure. The accurate identification of crop diseases is highly desired in the field of agricultural information. In this study, we propose a deep convolutional neural network (CNN)-based architecture (modified LeNet) for maize leaf disease classification. The experimentation is carried out using maize leaf images from the PlantVillage dataset. The proposed CNNs are trained to identify four different classes (three diseases and one healthy class). The learned model achieves an accuracy of 97.89%. The simulation results for the classification of maize leaf disease show the potential efficiency of the proposed method.

Maize leaf disease classification using deep convolutional neural networks

Segmentation is a crucial phase in image processing because it simplifies the representation of an image and facilitates its analysis. The multilevel thresholding method is more efficient for segmenting digital mammograms compared to the classic bi-level thresholding since it uses a higher number of intensities to represent different regions in the image. In the literature, there are different techniques for multilevel segmentation; however, most of these approaches do not obtain good segmented images. In addition, they are computationally expensive. Recently, statistical criteria such as Otsu, Kapur, and cross-entropy have been utilized in combination with evolutionary and swarm-based strategies to investigate the optimal threshold values for multilevel segmentation. In this paper, an efficient methodology for multilevel segmentation is proposed using the Harris Hawks Optimization (HHO) algorithm and the minimum cross-entropy as a fitness function. To substantiate the results and effectiveness of the HHO-based method, it has been tested over a benchmark set of reference images, with the Berkeley segmentation database, and with medical images of digital mammography. The proposed HHO-based solver is verified based on other comparable optimizers and two machine learning algorithms K-means and the Fuzzy IterAg. The comparisons were performed based on three groups. This first one is to provide evidence of the optimization capabilities of the HHO using the Wilcoxon test, and the second is to verify segmented image quality using the PSNR, SSIM, and FSIM metrics. Then, the third way is to verify the segmented image comparing it with the ground-truth through the metrics PRI, GCE, and VoI. The experimental results, which are validated by statistical analysis, show that the introduced method produces efficient and reliable results in terms of quality, consistency, and accuracy in comparison with the other methods. This HHO-based method presents an improvement over other segmentation approaches that are currently used in the literature.

An efficient Harris hawks-inspired image segmentation method

Cutting edge deep learning techniques allow for image segmentation with great speed and accuracy. However, application to problems in materials science is often difficult since these complex models may have difficultly learning meaningful image features that would enable extension to new datasets. In situ electron microscopy provides a clear platform for utilizing automated image analysis. In this work, we consider the case of studying coarsening dynamics in supported nanoparticles, which is important for understanding, for example, the degradation of industrial catalysts. By systematically studying dataset preparation, neural network architecture, and accuracy evaluation, we describe important considerations in applying deep learning to physical applications, where generalizable and convincing models are required. With a focus on unique challenges that arise in high-resolution images, we propose methods for optimizing performance of image segmentation using convolutional neural networks, critically examining the application of complex deep learning models in favor of motivating intentional process design.

/pdf/understanding-important-features-of-deep-learning-models-for-38jfs9onvh.pdf

Understanding important features of deep learning models for segmentation of high-resolution transmission electron microscopy images.

In this paper we propose a method for recognizing facial expressions using information from a pair of domains: one has labelled data and one with unlabelled data. As the two domains may differ in distribution, we depart from the traditional semi–supervised framework towards a transfer learning approach. In our method, which we call Annealed Label Transfer, the deep learner explores and predicts labels on the unsupervised part, yet, in order to prevent too much confidence in its predictions (as domains are not identical), the global error is regularized with a randomization input via an annealing process. The method’s evaluation is carried out on a set of four scenarios. The first two are standard benchmarks with expression faces in the wild, while the latter two have been little attempted before: face expression recognition in children and the study of the separability of anxiety-originated expressions in the wild. In all cases we show the superiority of the proposed method with respect to the strong baselines.

Annealed Label Transfer for Face Expression Recognition.

Recent years have shown that deep learned neural networks are a valuable tool in the field of computer vision. This paper addresses the use of two different kinds of network architectures, namely LeNet and Network in Network (NiN). They will be compared in terms of both performance and computational efficiency by addressing the classification and detection problems. In this paper, multiple databases will be used to test the networks. One of them contains images depicting burn wounds from pediatric cases, another one contains an extensive number of art images and other facial databases were used for facial keypoints detection.

The use of deep learning in image segmentation, classification and detection.

This papers presents the use of color imaging as a starting point of burn wound evaluation, by the discrimination between healthy skin and burn wound. The skin/burn area identification is performed pixel-wise, according to the properties of an entire encompassing patch. The classification is learned under a supervised scenario, according to a ground truth defined by specialist surgeons from a large pediatric case database, by a deep learned convolutional neural network. The database is extensive and was recorded over several months in real, hospital conditions The proposed approach achieves an overall performance comparable to the literature-reported average performance of a specialist surgeon.

Automatic burn area identification in color images

Humans are capable of perceiving a natural scene at a glance and painters, through their representations, push this capacity to the limit by abstraction Yet we still do it What about machines? In this paper we address the problem of recognizing the theme (scene type) in digitized paintings The approach is based on the Convolutional Neural Network framework and the chosen architecture is the recent Residual Network In the first level of evaluation, we determine the recognition rate of a CNN given a database of 80,000 annotated digitized paintings In the second level we evaluate the impact of extending the training database with photographs directly and via two domain adaptation functions and thus we are able to assess the abstraction level that CNN is capable to achieve

Domain Transfer for Delving into Deep Networks Capacity to De-Abstract Art

In this paper, as we aim to construct a semi-supervised learning algorithm, we exploit the characteristics of the Deep Convolutional Networks to provide, for an input image, both an embedding descriptor and a prediction. The unlabeled data is combined with the labeled one in order to provide synthetic data, which describes better the input space. The network is asked to provide a large margin between clusters, while new data is self-labeled by the distance to class centroids, in the embedding space. The method is tested on standard benchmarks for semi-supervised learning, where it matches state of the art performance and on the problem of face expression recognition where it increases the accuracy by a noticeable margin .

Mihai-Sorin Badea

Papers

Annealed Label Transfer for Face Expression Recognition.

The use of deep learning in image segmentation, classification and detection.

Automatic burn area identification in color images

Domain Transfer for Delving into Deep Networks Capacity to De-Abstract Art

Margin-Mix: Semi-Supervised Learning for Face Expression Recognition