We propose a deep learning approach for directly estimating relative atmospheric visibility from outdoor photos without relying on weather images or data that require expensive sensing or custom capture. Our data-driven approach capitalizes on a large collection of Internet images to learn rich scene and visibility varieties. The relative CNN–RNN coarse-to-fine model, where CNN stands for convolutional neural network and RNN stands for recurrent neural network, exploits the joint power of relative support vector machine, which has a good ranking representation, and the data-driven deep learning features derived from our novel CNN–RNN model. The CNN–RNN model makes use of shortcut connections to bridge a CNN module and an RNN coarse-to-fine module. The CNN captures the global view while the RNN simulates human’s attention shift, namely, from the whole image (global) to the farthest discerned region (local). The learned relative model can be adapted to predict absolute visibility in limited scenarios. Extensive experiments and comparisons are performed to verify our method. We have built an annotated dataset consisting of about 40000 images with 0.2 million human annotations. The large-scale, annotated visibility data set will be made available to accompany this paper.

Relative CNN-RNN: Learning Relative Atmospheric Visibility From Images

Small-object detection is a basic and challenging problem in computer vision tasks. It is widely used in pedestrian detection, traffic sign detection, and other fields. This paper proposes a deep learning small-object detection method based on image super-resolution to improve the speed and accuracy of small-object detection. First, we add a feature texture transfer (FTT) module at the input end to improve the image resolution at this end as well as to remove the noise in the image. Then, in the backbone network, using the Darknet53 framework, we use dense blocks to replace residual blocks to reduce the number of network structure parameters to avoid unnecessary calculations. Then, to make full use of the features of small targets in the image, the neck uses a combination of SPPnet and PANnet to complete this part of the multi-scale feature fusion work. Finally, the problem of image background and foreground imbalance is solved by adding the foreground and background balance loss function to the YOLOv4 loss function part. The results of the experiment conducted using our self-built dataset show that the proposed method has higher accuracy and speed compared with the currently available small-target detection methods.

/pdf/small-object-detection-based-on-yolo-and-dense-block-via-3pa9byhvrt.pdf

Small-Object Detection Based on YOLO and Dense Block via Image Super-Resolution

A review of state-of-the-art in face presentation attack detection: from early development to advanced deep learning and multi-modal fusion methods.

The unstructured data impacts 95% of the organizations and costs them millions of dollars annually. If managed well, it can significantly improve business productivity. The traditional information extraction techniques are limited in their functionality, but AI-based techniques can provide a better solution. A thorough investigation of AI-based techniques for automatic information extraction from unstructured documents is missing in the literature. The purpose of this Systematic Literature Review (SLR) is to recognize, and analyze research on the techniques used for automatic information extraction from unstructured documents and to provide directions for future research. The SLR guidelines proposed by Kitchenham and Charters were adhered to conduct a literature search on various databases between 2010 and 2020. We found that: 1. The existing information extraction techniques are template-based or rule-based, 2. The existing methods lack the capability to tackle complex document layouts in real-time situations such as invoices and purchase orders, 3. The datasets available publicly are task-specific and of low quality. Hence, there is a need to develop a new dataset that reflects real-world problems. Our SLR discovered that AI-based approaches have a strong potential to extract useful information from unstructured documents automatically. However, they face certain challenges in processing multiple layouts of the unstructured documents. Our SLR brings out conceptualization of a framework for construction of high-quality unstructured documents dataset with strong data validation techniques for automated information extraction. Our SLR also reveals a need for a close association between the businesses and researchers to handle various challenges of the unstructured data analysis.

/pdf/efficient-automated-processing-of-the-unstructured-documents-4nj8wruf94.pdf

Efficient Automated Processing of the Unstructured Documents Using Artificial Intelligence: A Systematic Literature Review and Future Directions

Image modalities, such as computed tomography (CT), magnetic resonance imaging (MRI), single‐photon emission computed tomography (SPECT), and so on, reflect various levels of details about objects of interest that help medical practitioners to examine patients' diseases from different perspectives. A single medical image, at times, may not be sufficient for making a critical decision; therefore, providing detailed information from a different perspective may help in making a better decision. Image fusion techniques play a vital role in this regard by combining important details from different medical images into a single, information enhanced image. In this article, we present a novel weighted term multimodality anatomical medical image fusion method. The proposed method, as a first step, eliminates the distortions from the source images and afterward, extracts two pieces of crucial information: the local contrast and the salient structure. Both the local contrast and salient structure are later combined to obtain the final weight map. The obtained weights are then passed through a fast guided filter to remove the discontinuities and noise. Lastly, the refined weight map is fused with source images using pyramid decomposition to get the final fused image. The proposed method is accessed and compared both qualitatively and quantitatively with state‐of‐the‐art techniques. The result illustrates the performance superiority and efficiency of the proposed method.

A novel multimodality anatomical image fusion method based on contrast and structure extraction

In order to solve the problem of face recognition in complex environments being vulnerable to illumination change, object rotation, occlusion, and so on, which leads to the imprecision of target position, a face recognition algorithm with multi-feature fusion is proposed. This study presents a new robust face-matching method named SR-CNN, combining the rotation-invariant texture feature (RITF) vector, the scale-invariant feature transform (SIFT) vector, and the convolution neural network (CNN). Furthermore, a graphics processing unit (GPU) is used to parallelize the model for an optimal computational performance. The Labeled Faces in the Wild (LFW) database and self-collection face database were selected for experiments. It turns out that the true positive rate is improved by 10.97–13.24% and the acceleration ratio (the ratio between central processing unit (CPU) operation time and GPU time) is 5–6 times for the LFW face database. For the self-collection, the true positive rate increased by 12.65–15.31%, and the acceleration ratio improved by a factor of 6–7.

Face Recognition Using the SR-CNN Model.

Medical image quality requirements have been increasingly stringent with the recent developments of medical technology. To meet clinical diagnosis needs, an effective medical image enhancement method based on convolutional neural networks (CNNs) and frequency band broadening (FBB) is proposed. Curvelet transform is used to deal with medical data by obtaining the curvelet coefficient in each scale and direction, and the generalised cross-validation is implemented to select the optimal threshold for performing denoising processing. Meanwhile, the cycle spinning scheme is used to wipe off the visible ringing effects along the edges of medical images. Then, FBB and a new CNN model based on the retinex model are used to improve the processed image resolution. Eventually, pixel-level fusion is made between two enhanced medical images from CNN and FBB. In the authors’ study, 50 groups of medical magnetic resonance imaging, X-ray, and computed tomography images in total have been studied. The experimental results indicate that the final enhanced image using the proposed method outperforms other methods. The resolution and the edge details of the processed image are significantly enhanced, providing a more effective and accurate basis for medical workers to diagnose diseases.

https://ietresearch.onlinelibrary.wiley.com/doi/epdf/10.1049/iet-ipr.2018.6380

Efficient medical image enhancement based on CNN-FBB model

The identification which uses biological characteristics has been a current top in the recent past. However, numerous spoofing skills occur with the rising prosperity of advance recognition technology, especially in the detection and recognition of a face. In allusion to the problem above, more robust and accurate face spoofing detection schemes have been put forward. Convolutional neural networks (CNNs) have demonstrated extraordinary success in face liveness detection recently. In this study, an effective face anti-spoofing detection method based on CNN and rotation invariant local binary patterns (RI-LBP) has been proposed. First, the authors use CNN to extract deep features and use RI-LBP to extract colour texture features. In addition, the principal component analysis approach is employed to decrease the dimensions of deep characteristic. Moreover, two different features are fused before applying to support vector machine (SVM). Finally, the SVM classifier is adopted to identify genuine faces from fake faces. They have conducted extensive experiments to obtain a scheme of better generalisation capability for face anti-spoofing detection. The analysis results indicate that the proposed approach implements great generalisation capability over other state-of-the-art approaches within the intra-databases and cross-databases.

Face liveness detection: fusing colour texture feature and deep feature

Hospitals need to invest a lot of manpower to manually input the contents of medical invoices (nearly 300,000,000 medical invoices a year) into the medical system. In order to help the hospital save money and stabilize work efficiency, this paper designed a system to complete the complicated work using a Gaussian blur and smoothing-convolutional neural network combined with a recurrent neural network (GBS-CR) method. Gaussian blur and smoothing (GBS) is a novel preprocessing method that can fix the breakpoint font in medical invoices. The combination of convolutional neural network (CNN) and recurrent neural network (RNN) was used to raise the recognition rate of the breakpoint font in medical invoices. RNN was designed to be the semantic revision module. In the aspect of image preprocessing, Gaussian blur and smoothing were used to fix the breakpoint font. In the period of making the self-built dataset, a certain proportion of the breakpoint font (the font of breakpoint is 3, the original font is 7) was added, in this paper, so as to optimize the Alexnet-Adam-CNN (AA-CNN) model, which is more suitable for the recognition of the breakpoint font than the traditional CNN model. In terms of the identification methods, we not only adopted the optimized AA-CNN for identification, but also combined RNN to carry out the semantic revisions of the identified results of CNN, meanwhile further improving the recognition rate of the medical invoices. The experimental results show that compared with the state-of-art invoice recognition method, the method presented in this paper has an average increase of 10 to 15 percentage points in recognition rate.

/pdf/dual-model-medical-invoices-recognition-79izo6x6fp.pdf

Dual Model Medical Invoices Recognition

The invention provides a label box-free micro-seismic signal detection method and device. The method comprises the following steps: two data sets a and b containing effective signals are screened outfrom micro-seismic signal data, and preprocessing is carried out; image feature extraction is carried out through a convolutional neural network; by means of an RPN layer in a Faster-RCNN, generatingcandidate boxes on data feature graphs; discriminating the similarity of the candidate boxes by a discriminator of a generative adversarial network, thereby obtaining the candidate boxes with the similarity exceeding a preset threshold; adjusting the candidate boxes through ROI Align to generate feature graphs with the same size; after passing through a full-connection layer of the Faster-RCNN, classifying the feature graphs with the same size, analyzing a prediction error by using a cross entropy loss function, and performing continuous fitting and training to obtain a label box-free micro-seismic signal detection model; and inputting a to-be-detected signal image, and detecting the effective signals through the label box-free micro-seismic signal detection model. The micro-seismic signaldetection can be realized through the label box-free data sets, so that the detection precision is improved.

Tang Xingong

Papers

Face Recognition Using the SR-CNN Model.

Efficient medical image enhancement based on CNN-FBB model

Face liveness detection: fusing colour texture feature and deep feature

Dual Model Medical Invoices Recognition

Label box-free micro-seismic signal detection method and device