Intelligent Interpretation Research Articles

PGNet: Positioning Guidance Network for Semantic Segmentation of Very-High-Resolution Remote Sensing Images

Semantic segmentation of very-high-resolution (VHR) remote sensing images plays an important role in the intelligent interpretation of remote sensing since it predicts pixel-level labels to the images. Although many semantic segmentation methods of VHR remote sensing images have emerged recently and achieved good results, it is still a challenging task because the objects of VHR remote sensing images show large intra-class and small inter-class variations, and their size varies in a large range. Therefore, we proposed a novel semantic segmentation framework for VHR remote sensing images, called Positioning Guidance Network (PGNet), which consists of the feature extractor, a positioning guiding module (PGM), and a self-multiscale collection module (SMCM). First, the PGM can extract long-range dependence and global context information with the help of the transformer architecture and effectively transfer them to each pyramid-level feature, thus effectively improving the segmentation effectiveness between different semantic objects. Secondly, the SMCM we designed can effectively extract multi-scale information and generate high-resolution feature maps with high-level semantic information, thus helping to segment objects in small and varying sizes. Without bells and whistles, the mIoU scores of the proposed PGNet on the iSAID dataset and ISPRS Vaihingn dataset are 1.49% and 2.40% higher than FactSeg, respectively.

A multi-branch hierarchical attention network for medical target segmentation

Medical imaging techniques have been widely used in modern clinical disease diagnosis and treatment programs. The captured medical images can well reflect the conditions of the human body tissues, which are significantly helpful to the doctors to determine the existence or the severity of the disease. In this paper, we develop a hierarchical attentive high-resolution convolutional network (AttHRNet) for segmenting targets of interest from medical images aiming to improve the automated processing standard and the intelligent interpretation quality of the medical images. The AttHRNet is an improved version of the high-resolution network (HRNet) structure with three novel modules. First, built with an improved HRNet structure assisted by a multiscale context augmentation (MSCA) module as the feature extraction backbone, the AttHRNet can produce a set of high-quality, strong-semantic feature maps at different resolutions. The MSCA module functions to reduce the information loss during feature downsampling. Second, designed with an effective feature attention principle, the feature encoding quality in each branch can be significantly promoted by concentrating on the informative and salient feature encodings across both channels and spatial locations. Furthermore, formulated with a hierarchical segmentation scheme, the output feature maps can be further augmented by including the semantic-level category exploitation (SLCE) module with a global perspective. The SLCE module allows the information from lower resolution segmentations to inform higher resolution segmentations. Through quantitative examinations, visual verifications, and comparative evaluations on four medical image datasets, we convince the promising applicability and competitive superiority of the AttHRNet in medical target segmentation issues.

Monitoring and automatic characterization of cracks in strain-hardening cementitious composite (SHCC) through intelligent interpretation of photos

This paper presents an intelligent photo interpretation approach to automatically monitor and characterize dense interconnected microcracks in strain-hardening cementitious composite (SHCC) featuring unique crack patterns in terms of crack number and crack width. The presented approach employs a stereo vision system that integrates binocular and monocular cameras for automatic detection, ranging, and quantification of cracks as well as characterization of crack patterns. The presented approach was implemented into evaluation of SHCC in flexural tests and direct tension tests. Dense microcracks were detected and ranged by the stereo vision system, segmented by an encoder-decoder approach, and quantified by an efficient computer vision approach. Evolution of the cracks was traced throughout the loading process until failure, and a statistical analysis revealed that the crack width was retained while the crack number monotonically increased. The interpretation time was shorter than 0.4 s for each photo, making the approach promising for monitoring of SHCC. The proposed system can be deployed for automated assessment of cementitious composites with complex crack patterns in material research and engineering structures.

Edge-aware and spectral–spatial information aggregation network for multispectral image semantic segmentation

Semantic segmentation is a fundamental task in the field of remote sensing image intelligent interpretation and computer vision. Multispectral remote sensing images have attracted more and more researchers’ attention because they can accurately describe different types of reflection spectra. However, inaccurate multispectral feature description leads to edge semantic ambiguity and misclassification of small objects. In this article, we propose a novel network named edge-aware and spectral–spatial information aggregation net (ESSANet) to capture both high-level semantic features and low-level edge details for semantic segmentation of remote sensing images. Specifically, on the one hand, in order to improve the representation ability of discriminant features, we design a two-stream spectral–spatial feature extraction network via 3D hybrid convolution and multi-level aggregation network. On the other hand, in order to eliminate the effect of edge semantic ambiguity, we develop a siamese edge-aware structure and multi-stage edge loss function. Experimental results show that our method achieved 3.5% and 4.09% mean intersection over union (mIoU) score improvements and 2.59% and 3.32% Kappa score improvements compared with the competitive baseline algorithm on the SEN12MS and US3D datasets, respectively. In addition, the method proposed in this paper also achieves a better trade-off between speed and accuracy.

HrreNet: semantic segmentation network for moderate and high-resolution satellite images

ABSTRACT Semantic segmentation is a commonly used intelligent interpretation method. However, semantic segmentation via deep learning usually faces several problems: resolution loss due to downsampling, difficulty in obtaining context information, and imprecise object boundaries. These problems lead to a poor performance when using satellite images instead of conventional images and limit the value of intelligent interpretation of satellite images in practical application. In this study, we construct a Qilian Ecological Resource Extraction Dataset (QERED) with multiple scene satellite images of Qilian after preprocessing via correction, fusion, mosaic, and annotation. We then propose a segmentation network, namely, High-resolution Resource Extracting Network (HrreNet), by using high-resolution feature representation, multi-scale context fusion, boundary refinement with relearning, and structural similarity loss. Experiments on QERED show that HrreNet greatly improves performance on small-size objects and slightly improves performance on larger-size objects. HrreNet achieves the best result of 73.36% mIoU on a moderate-resolution satellite dataset CCFD and 80.87% mIoU on a high-resolution satellite dataset Vaihingen.

DEEP LEARNING TRAINING WITH UNBALANCE SAMPLE DISTRIBUTION FOR REMOTE SENSING IMAGE SEGMENTATION

Abstract. The intelligent interpretation of remote sensing images based on deep learning has become a hot spot with the increasing satellite images acquired due to the rapid development of aerospace technology. Sufficient and reasonable distributed samples are essential for the accuracy of deep learning. The spatial distribution of natural features is inhomogeneous in the real world. When people create sample dataset, they often collect within a certain local range, which may bring problems of unbalanced distribution of samples, including the unbalance between training dataset and validation dataset, and the unbalance among different sample categories. This long-tail distribution of samples (i.e., a few classes account for most of the data, while most classes are under-represented) can lead to bias in the training model and make it difficult to ensure accuracy.In this paper we tried to solved the above-mentioned problem in landcover classification with high spatial and spectral resolution (HSSR) remote sensing images. We first adopted an iterative stratification method for multi-label data classification to ensure that both training dataset and validation dataset contain reasonable proportion of landcover classes. Then we proposed a weighted loss algorithm to further strengthen the learning ability of the model for rare categories. Experiments on a large volume HSSR dataset shows that with our methods the accuracy of landcover classification increased by 2%.

ATA2022 Annual Conference & Expo May 1–3, 2022Boston, Massachusetts

ATA2022 Annual Conference & Expo May 1–3, 2022Boston, Massachusetts

Why PGT-A, most likely, improves IVF success

Preimplantation genetic testing for aneuploidies (PGT-A), with its vocal advocates and opponents, is at the epicentre of a perpetual, often heated, debate. The main issues include the following. First, how do we interpret the existing evidence-base? Around 100 retrospective and single-centre studies, two non-selection trials and at least two meta-analyses point to its efficacy in improving live birth rates, although randomized controlled trials are more mixed. Second, what should be done in relation to euploid/aneuploid mosaicism? Recent data suggest that low-level mosaic pregnancies can proceed uneventfully to term, so intelligent interpretation of the diagnostic data is appropriate. Third, what is the stance of the Human Fertilisation and Embryology Authority? The ‘traffic light’ system is much debated and is perhaps best described as well-intentioned, but misguided in places. Fourth, what is the motivation of people who maintain their point of view despite the evidence? Sadly, the presentation of new empirical evidence polarizes, rather than reconciles, opinion. Too many have made a career out of either promoting or denigrating PGT-A for them to back down easily. Finally, how can we find common ground and move forward? All patients should be counselled in a non-directive manner on whether to embark on PGT-A, summarizing for them the whole evidence base so they can make up their own mind.

Role of Artificial Intelligence Interpretation of Colposcopic Images in Cervical Cancer Screening.

The accuracy of colposcopic diagnosis depends on the skill and proficiency of physicians. This study evaluated the feasibility of interpreting colposcopic images with the assistance of artificial intelligence (AI) for the diagnosis of high-grade cervical intraepithelial lesions. This study included female patients who underwent colposcopy-guided biopsy in 2020 at two institutions in the Republic of Korea. Two experienced colposcopists reviewed all images separately. The Cerviray AI® system (AIDOT, Seoul, Korea) was used to interpret the cervical images. AI demonstrated improved sensitivity with comparable specificity and positive predictive value when compared with the colposcopic impressions of each clinician. The areas under the curve were greater with combined impressions (both AI and that of the two colposcopists) of high-grade lesions, when compared with the individual impressions of each colposcopist. This study highlights the feasibility of the application of an AI system in cervical cancer screening. AI interpretation can be utilized as an assisting tool in combination with human colposcopic evaluation of exocervix.

A Retrospective Clinical Evaluation of an Artificial Intelligence Screening Method for Early Detection of STEMI in the Emergency Department.

BackgroundRapid revascularization is the key to better patient outcomes in ST-elevation myocardial infarction (STEMI). Direct activation of cardiac catheterization laboratory (CCL) using artificial intelligence (AI) interpretation of initial electrocardiography (ECG) might help reduce door-to-balloon (D2B) time. To prove that this approach is feasible and beneficial, we assessed the non-inferiority of such a process over conventional evaluation and estimated its clinical benefits, including a reduction in D2B time, medical cost, and 1-year mortality.MethodsThis is a single-center retrospective study of emergency department (ED) patients suspected of having STEMI from January 2021 to June 2021. Quantitative ECG (QCG™), a comprehensive cardiovascular evaluation system, was used for screening. The non-inferiority of the AI-driven CCL activation over joint clinical evaluation by emergency physicians and cardiologists was tested using a 5% non-inferiority margin.ResultsEighty patients (STEMI, 54 patients [67.5%]) were analyzed. The area under the curve of QCG score was 0.947. Binned at 50 (binary QCG), the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were 98.1% (95% confidence interval [CI], 94.6%, 100.0%), 76.9% (95% CI, 60.7%, 93.1%), 89.8% (95% CI, 82.1%, 97.5%) and 95.2% (95% CI, 86.1%, 100.0%), respectively. The difference in sensitivity and specificity between binary QCG and the joint clinical decision was 3.7% (95% CI, −3.5%, 10.9%) and 19.2% (95% CI, −4.7%, 43.1%), respectively, confirming the non-inferiority. The estimated median reduction in D2B time, evaluation cost, and the relative risk of 1-year mortality were 11.0 minutes (interquartile range [IQR], 7.3–20.0 minutes), 26,902.2 KRW (22.78 USD) per STEMI patient, and 12.39% (IQR, 7.51–22.54%), respectively.ConclusionAI-assisted CCL activation using initial ECG is feasible. If such a policy is implemented, it would be reasonable to expect some reduction in D2B time, medical cost, and 1-year mortality.

Innovation of French Interpretation Teaching Mode in the New Liberal Arts Environment.

The status of French as a language has improved to some extent in recent years as a result of the exchange and blending of cultures from various nations. Each language, including French, must undergo a protracted process in order to become more widely spoken. It is essential to develop top-notch French interpreters in order to lower the barriers preventing cross-border communication. The teaching of French interpretation has recently received more attention in many colleges, but the issue of a single teaching method has long persisted. In recent years, a new approach to teaching the liberal arts has emerged. Its main goal is to implement comprehensive interdisciplinary teaching by fusing contemporary information technology with traditional liberal arts instruction. This paper conducted a cutting-edge study on the teaching method of French interpretation against the backdrop of the new liberal arts in order to alter the teaching strategy and increase teaching effectiveness. In this paper, an intelligent interpretation teaching system was constructed using information technology. Through text classification algorithms, it incorporated and optimized teaching resources. According to the experimental findings, the teaching process was optimized, and the teaching efficiency increased by 7.94% when French interpretation was taught against a backdrop of new liberal arts.

Background Representation Learning With Structural Constraint for Hyperspectral Anomaly Detection

Hyperspectral anomaly detection is a popular topic in remote sensing image intelligent interpretation. To detect anomaly, many methods for background representation have been proposed. However, the prior information of background and anomaly is not fully explored in these methods. To tackle this issue, we combine low-rank dictionary learning (LRDL) with total variation (TV) constraint for hyperspectral anomaly detection. To be specific, the LRDL is introduced for background representation to explore the low-rank priori of background. Considering the smooth structural characteristic of background in spatial, we introduce the TV constraint on coefficients matrix for better background representation learning. Then the residual part is used to discriminate anomaly. The experiments on three real data sets demonstrate the effectiveness of the proposed method compared with state-of-the-art algorithms in hyperspectral anomaly detection.

Adversarial Deception Against SAR Target Recognition Network

Synthetic Aperture Radar (SAR) automatic target recognition (ATR) technology is one of the key technologies to achieve intelligent interpretation for SAR images. With the rapid development of deep learning, deep neural networks have been successively used in SAR ATR and show priority in comparison with the conventional methods. Recently, more and more attention is paid to the robustness of deep learning based SAR ATR methods. The reason is that maliciously modified and imperceptible adversarial images can deceive the SAR ATR methods which are based on the deep neural networks. In this paper, we propose a novel SAR ATR adversarial deception algorithm, which fully considers the characteristics of SAR data. Our method can obtain the satisfactory perturbations with a higher deception success rate, a higher recognition confidence, and a smaller perturbation coverage than other state-of-the- art methods for the SAR images. Experimental results using the MSTAR dataset and OpenSARShip dataset demonstrate the effectiveness of our method. The proposed adversarial deception method can be used in the applications such as SAR dataset protection, SAR sensor design and SAR image quality evaluation.

Fused Adaptive Receptive Field Mechanism and Dynamic Multiscale Dilated Convolution for Side-Scan Sonar Image Segmentation

Side-scan sonar (SSS) is a vital sensor for marine survey, which is widely used in military and civilian fields. The accurate segmentation of SSS images is critical in sonar image intelligent interpretation. Existing SSS image segmentation methods have several limitations, such as insufficient feature extraction, relatively worse segmentation results for tiny target categories, and serious interference by seabed reverberation noise and bright shadow region. To overcome these issues, we propose a novel encoder-decoder architecture SSS image segmentation method based on convolution neural network (CNN). First, we extract the multi-scale feature information contained in target region using the dynamic multi-scale dilated convolution (DMDC_Conv). Second, to further obtain the global and detail feature information, we construct the adaptive receptive field mechanism block (ARFM_Block). Third, we design a feature fusion attention mechanism block (FFAM_Block) to fuse high-level and low-level feature information with different scales and suppress background information interference. Final, we construct a tree structure optimization module (TSOM) to solve the problem of pixel misclassification and obtain refine SSS image segmentation results. Extensive experiments are carried out on the constructed real scene SSS image dataset. The experimental results show that the proposed method achieves 93.24% and 90.82% of MPA and MIoU, respectively, which outperforms other state-of-the-art methods and has a substantial advantage in inference speed and calculation parameters.

Development of Literature Database of Korean Ritual Studies and Intelligent Platform

The Korean Ritual Studies Series is a compilation of the works about Korean ritual studies, which makes a great contribution to scholarship in this area. Relying on a number of major national programs in China and South Korea, digital collation and research the Korean Ritual Studies Series was made to create into a “literature database of Korean ritual studies” by cooperation between the Center for the Study and Application of Chinese Characters (CSACC) of East China Normal University (ECNU) which is a key research institute of humanities and social science approved by the Ministry of Education of the People’s Republic of China, and the research institutions of Kyungsung University including the Center for the Study of Chinese Characters in Korea, the Research Society of Korean Ritual Studies and the Academy of Korean Studies. The basic framework of the database contains six sub-databases, which are fully correlated with each other according to relevant attributes. The database provides search capability, with three search functions, including catalog index, full-text retrieval and a word index. It is a major achievement of an international cooperation program supported by a multilevel international platform. It provides a large sustainable development space for the “literature database of Korean ritual studies”, which is an indispensable and important part of the “public database of Chinese characters” created by the Center for the Study and Application of Chinese Characters (CSACC) of East China Normal University (ECNU). The three key breakthroughs are needed for its further development: (I) Achieving the data association, comparative research and web publishing of Chinese and Korean ritual literature, establishing an Asian big data ritual research and application platform, supporting the research of Chinese and Korean ritual cultures, the communication history of ritual culture, the development history of Chinese characters and the research of many other disciplines, and meeting the multi-functional needs of ritual research, teaching and communication while completing a comprehensive digital connection of the internal resources of the database. (II) Deeply tapping into the massive Chinese character writing and memory data to create a 21st century database for written texts in the Sinosphere on the basis of digital processing and research of Korean ritual literature. (III) Realizing the multidimensional intelligent automatic recognition and interpretation of original literature written and memorized in Chinese characters such as ancient Chinese and Korean ritual literature, family mottoes and letters of noble families and great clans as well as folk handcopied books, and opening a new era of intelligent interpretation and research of the written texts in the Sinosphere through image recognition technology.

Remote-Sensing Image Tile-Level Annotation Based on Discriminative Features and Expressive Visual Word Descriptors

Remote-sensing images play a crucial role in a wide range of applications and have been receiving significant attention. In recent years, great efforts have been made in developing various methods for intelligent interpretation of remote-sensing images. Generally speaking, machine learning-based methods of remote-sensing image interpretation require a large number of labeled samples and there are still not enough annotated datasets in the field of remote sensing. However, manual annotation of remote-sensing images is usually labor-intensive and requires expert knowledge and the accuracy of annotation results is relatively low. The goal of this paper is to propose a novel tile-level annotation method of remote-sensing images to obtain remote-sensing datasets which are well-labeled and contain accurate semantic concepts. Firstly, we use a set of images with defined semantic concepts to represent the training set and divide them into several nonoverlapping regions. Secondly, the color features, texture features, and spatial features of each region are extracted, and discriminative features are obtained by the weight optimization feature fusion method. Then, the features are quantized into visual words by applying a density-based clustering center selection method and an isolated feature point elimination method. And the remote-sensing images can be represented by a series of visual words. Finally, the LDA model is used to calculate the probabilities of semantic categories for each region. The experiments are conducted on remote-sensing images which demonstrate that our proposed method can achieve good performance on remote-sensing image tile-level annotation. The implications of our research can obtain annotated datasets with accurate semantic concepts for intelligent interpretation of remote-sensing images.

A rapid and affordable point of care test for antibodies against SARS-CoV-2 based on hemagglutination and artificial intelligence interpretation

Diagnostic tests that detect antibodies (AB) against SARS-CoV-2 for evaluation of seroprevalence and guidance of health care measures are important tools for managing the COVID-19 pandemic. Current tests have certain limitations with regard to turnaround time, costs and availability, particularly in point-of-care (POC) settings. We established a hemagglutination-based AB test that is based on bi-specific proteins which contain a dromedary-derived antibody (nanobody) binding red blood cells (RBD) and a SARS-CoV-2-derived antigen, such as the receptor-binding domain of the Spike protein (Spike-RBD). While the nanobody mediates swift binding to RBC, the antigen moiety directs instantaneous, visually apparent hemagglutination in the presence of SARS-CoV-2-specific AB generated in COVID-19 patients or vaccinated individuals. Method comparison studies with assays cleared by emergency use authorization demonstrate high specificity and sensitivity. To further increase objectivity of test interpretation, we developed an image analysis tool based on digital image acquisition (via a cell phone) and a machine learning algorithm based on defined sample-training and -validation datasets. Preliminary data, including a small clinical study, provides proof of principle for test performance in a POC setting. Together, the data support the interpretation that this AB test format, which we refer to as ‘NanoSpot.ai’, is suitable for POC testing, can be manufactured at very low costs and, based on its generic mode of action, can likely be adapted to a variety of other pathogens.

General intelligence disentangled via a generality metric for natural and artificial intelligence

Success in all sorts of situations is the most classical interpretation of general intelligence. Under limited resources, however, the capability of an agent must necessarily be limited too, and generality needs to be understood as comprehensive performance up to a level of difficulty. The degree of generality then refers to the way an agent’s capability is distributed as a function of task difficulty. This dissects the notion of general intelligence into two non-populational measures, generality and capability, which we apply to individuals and groups of humans, other animals and AI systems, on several cognitive and perceptual tests. Our results indicate that generality and capability can decouple at the individual level: very specialised agents can show high capability and vice versa. The metrics also decouple at the population level, and we rarely see diminishing returns in generality for those groups of high capability. We relate the individual measure of generality to traditional notions of general intelligence and cognitive efficiency in humans, collectives, non-human animals and machines. The choice of the difficulty function now plays a prominent role in this new conception of generality, which brings a quantitative tool for shedding light on long-standing questions about the evolution of general intelligence and the evaluation of progress in Artificial General Intelligence.

Logical and conceptual constructions, theoretical and legal versions of the interpretation of the artificial intelligence’s "legal personality"

The lack of sufficient legal regulation of relations associated with the use of artificial intelligence and robotics is associated with the conceptual lack of resolution of the basic issues underlying modern models of legal regulation for anything: about the concept and general status of the phenomenon to be settled. With regard to the research topic, this translates into questions about the concept and legal personality of artificial intelligence. These questions represent a kind of logical metaphor: in order to start modeling the legal personality of artificial intelligence, we must first define its concept; and the legal concept of artificial intelligence will be formulated differently depending on whether we consider it as a subject or an object of law. The authors investigate the technical, moral and legal aspects of the problem concerning the status and understanding artificial intelligence, and draw conclusions about the basic conditions for the legal regulation of this area.

A Practical Method for the Automatic Recognition of Rock Structures in Panoramic Borehole Image during Deep-Hole Drilling Engineering

Digital panoramic borehole imaging technology has been widely used in the practice of drilling engineering. Based on many high-definition panoramic borehole images obtained by the borehole imaging system, this paper puts forward an automatic recognition method based on clustering and characteristic functions to perform intelligent analysis and automatic interpretation researches, and successfully applied to the analysis of the borehole images obtained at the Wudongde Hydropower Station in the south-west of China. The results show that the automatic recognition method can fully and quickly automatically identify most of the important structural planes and their position, dip, dip angle and gap width and other characteristic parameter information in the entire borehole image. The recognition rate of the main structural plane is about 90%. The accuracy rate is about 85%, the total time cost is about 3 h, and the accuracy deviation is less than 4% among the 12 boreholes with a depth of about 50 m. The application of automatic recognition technology to the panoramic borehole image can greatly improve work efficiency, reduce the time cost, and avoid the interference caused by humans, making it possible to automatically recognize the structural plane parameters of the full-hole image.