The measurement matrix which plays an important role in compressed sensing has got a lot of attention. However, the existing measurement matrices ignore the energy concentration characteristic of the natural images in the sparse domain, which can help to improve the sensing efficiency and the construction efficiency. Here, the authors propose a simple but efficient measurement matrix based on the Hadamard matrix, named Hadamard-diagonal matrix (HDM). In HDM, the energy conservation in the sparse domain is maximised. In addition, considering the reconstruction performance can be further improved by decreasing the mutual coherence of the measurement matrix, an effective optimisation strategy is adopted in order to reducing the mutual coherence for better reconstruction quality. The authors conduct several experiments to evaluate the performance of HDM and the effectiveness of optimisation algorithm. The experimental results show that HDM performs better than other popular measurement matrices, and the optimisation algorithm can improve the performance of not only the HDM but also the other popular measurement matrices.

Saliency map estimation in computer vision aims to estimate the locations where people gaze in images. Since people tend to look at objects in images, the parameters of the model pre-trained on ImageNet for image classification are useful for the saliency map estimation. However, there is no research on the relationship between the image classification accuracy and the performance of the saliency map estimation. In this study, it is shown that there is a strong correlation between image classification accuracy and saliency map estimation accuracy. The authors also investigated the effective architecture based on multi-scale images and the up-sampling layers to refine the saliency-map resolution. The model achieved the state-of-the-art accuracy on the PASCAL-S, OSIE, and MIT1003 datasets. In the MIT saliency benchmark, the model achieved the best performance in some metrics and competitive results in the other metrics.

Annotations are critical for machine learning and developing computer aided diagnosis (CAD) algorithms. Good performance of CAD is critical to their adoption, which generally rely on training with a wide variety of annotated data. However, a vast amount of medical data is either unlabeled or annotated only at the image-level. This poses a problem for exploring data driven approaches like deep learning for CAD. In this paper, we propose a novel crowdsourcing and synthetic image generation for training deep neural net-based lesion detection. The noisy nature of crowdsourced annotations is overcome by assigning a reliability factor for crowd subjects based on their performance and requiring region of interest markings from the crowd. A generative adversarial network-based solution is proposed to generate synthetic images with lesions to control the overall severity level of the disease. We demonstrate the reliability of the crowdsourced annotations and synthetic images by presenting a solution for training the deep neural network (DNN) with data drawn from a heterogeneous mixture of annotations. Experimental results obtained for hard exudate detection from retinal images show that training with refined crowdsourced data/synthetic images is effective as detection performance in terms of sensitivity improves by 25%/27% over training with just expert-markings.

To effectively mine the contents embedded in web images, it is useful to classify the images into different types so that they can be fed to different procedures for detailed analysis. The authors herein propose a hierarchical algorithm for efficiently classifying web images into four classes. Their algorithm consists of two stages: the first stage extracts global features reflecting the distributions of color, edge and gradient, and uses a support vector machine (SVM) classifier for preliminary classification. Images assigned low confidence by the first stage classifier are processed by the second stage, which further extracts local texture features represented in the bag-of-words framework and uses another SVM classifier for final classification. In addition, they design two fusion strategies to train the second-stage classifier and generate the final prediction depending on the usage of local features in the second stage. To validate the effectiveness of proposed method, they built a database containing more than 55,000 images from various sources. On their test image set, they obtained an overall classification accuracy of 98.4% and the processing speed is over 27 fps on an Intel(R) Xeon(R) central processing unit (2.90 GHz).

Achieving good recognition results for License plates is challenging due to multiple adverse factors. For instance, in Malaysia, where private vehicle (e.g., cars) have numbers with dark background, while public vehicle (taxis/cabs) have numbers with white background. To reduce the complexity of the problem, we propose to classify the above two types of images such that one can choose an appropriate method to achieve better results. Therefore, in this work, we explore the combination of Convolutional Neural Networks (CNN) and Recurrent Neural Networks namely, BLSTM (Bi-Directional Long Short Term Memory), for recognition. The CNN has been used for feature extraction as it has high discriminative ability, at the same time, BLSTM has the ability to extract context information based on the past information. For classification, we propose Dense Cluster based Voting (DCV), which separates foreground and background for successful classification of private and public. Experimental results on live data given by MIMOS, which is funded by Malaysian Government and the standard dataset UCSD show that the proposed classification outperforms the existing methods. In addition, the recognition results show that the recognition performance improves significantly after classification compared to before classification.

Achieving high recognition rate for license plate images is challenging due to multi-type images. We present new symmetry features based on stroke width for classifying each input license image as private, taxi, cursive text, when they expand the symbols by writing and non-text such that an appropriate optical character recognition (OCR) can be chosen for enhancing recognition performance. The proposed method explores gradient vector flow (GVF) for defining symmetry features, namely, GVF opposite direction, stroke width distance, and stroke pixel direction. Stroke pixels in Canny and Sobel which satisfy the above symmetry features are called local candidate stroke pixels. Common stroke pixels of the local candidate stroke pixels are considered as the global candidate stroke pixels. Spatial distribution of stroke pixels in local and global symmetry are explored by generating a weighted proximity matrix to extract statistical features, namely, mean, standard deviation, median and standard deviation with respect the median. The feature matrix is finally fed to an support vector machine (SVM) classifier for classification. Experimental results on large datasets for classification show that the proposed method outperforms the existing methods. The usefulness and effectiveness of the proposed classification is demonstrated by conducting recognition experiments before and after classification.