Picture Dataset Research Articles

Pengenalan Tulisan Tangan Huruf Latin Bersambung Menggunakan Local Binary Pattern dan K-Nearest Neighbor

There are 26 Latin letters in Indonesia, 5 of which are vowels and 21 consonants. This study will translate handwriting with a Latin object using the K-Nearest Neighbor method with the Local Binary Pattern extension. The research is being done with a focus on experimentation using a few methods that have already been discussed. Concatenated Latin letters have a few variations that depend on the work's author, so research will be conducted to identify cursive Latin letters based on these variations. Each of the 30 respondents wrote 26 capital letters and 26 lowercase letters on paper, which was then scanned to provide the image data. Black, blue, and red pens were used to write by every ten responders. The recognition procedure is broken into two halves, capital and non-capital letter recognition using 780 picture datasets each. In the study, k-fold cross-validation is used, with k = 6. The best value was reached at k = 7 with 29.49 percent accuracy, 33.88 percent precision, recall 33.46 percent, and F1-score 27.65 percent according to the research utilizing KNN with values k = 3, 5, and 7. and for recognizing non-capital characters, the best result was found at k=3 with accuracy, precision, recall, and F1-score of 26.28, 27.27, and 22.7%, respectively.

Detection of microcracks in silicon solar cells using Otsu-Canny edge detection algorithm

The necessity for photovoltaic (PV) arrays has expanded due to the rising demand for solar electrical energy. The demand for solar cells has grown as they are a key component of the PV array. Recent years have seen a rise in solar cell production as a result of this demand. The quality of a silicon panel affects its lifespan and effectiveness in the production of solar power. The picture edge-detection method is regularly employed to identify silicon solar panel flaws. On the other hand, defect identification is impacted by the panel's grid shadow. The most modern defect detection technique uses artificial neural networks, and pre-training the model requires a large number of potential samples from picture data sets. This paper offers the innovative Otsu-Canny operator method as a fault discovery strategy to address this problem. Rapid and accurate identification of microcracks enhances the effectiveness of fault detection. The Otsu algorithm is used in this method to improve edge detection performance for microcracks by modifying the canny operator with a double threshold and eliminating the impact of grids on detection. The process was finished by performing noise removal with morphological operation, picture segmentation, and image binarization. The experimental findings demonstrate a significant improvement in purity and integrity of the image processing algorithm used in this research. The Otsu-Canny edge detection technique is more accurate than other conventional edge detection algorithms for detecting microcracks, with a 92.83 % detection rate.

IMPLEMENTASI ARTIFICIAL INTELLIGENCE PADA PENGEMBANGAN SISTEM DETEKSI LIMBAH INDUSTRI PRODUK KULIT OTOMATIS

Rapid globalization has led to the growth of a significant amount of waste, including in the industry. Due to the considerable amount of growth of waste every year, effective and efficient waste management is needed to protect our environment. In the leather product industry, waste management is strongly important since it may have a significant impact to the employee and production process. Regarding those issues, waste management technology is considered proposed in order to solve those problems. Current research reported the outstanding work of implementing artificial intelligence for detecting and recognizing industrial waste. Artificial intelligence was proven to be a highly recommended approach that is able to classify several waste types with outstanding performance. Regardless of those facts, artificial intelligence still remains several hurdles, such as the high computational demands, especially for deep learning networks. Regarding the mentioned issue, we proposed a more proper deep learning network for recognizing industrial waste. In this research work, we use Single Shot Detector (SSD) to recognize and classify industrial waste. Our proposed solution was performed in the TrashNet dataset and Waste Picture dataset. Our proposed solution successfully achieve mAP of 0.8813, accuracy of 0,9795, precision of 0,9985 and recall of 0,9693 in the training process. Whereas, in the testing process we achieve average accuracy of 0,8254. According to those results, we can conclude that our proposed solution is suitable for industrial waste detection and has potential to be implemented as an embedded system for recognizing industrial waste automatically.

Modelling of deep learning enabled lung disease detection and classification on chest X-ray images

ABSTRACT The survival rate of people who use medical imaging modalities is improved by earlier detection of lung illness. The most recent developments in deep learning (DL) make it possible to correctly recognize and categorize lung disorders in various medical images. To categorize diseases in a publicly available chest X-ray picture dataset, a variety of machine learning (ML), deep learning (DL), and transfer learning algorithms have been explored in the current study. This research offers an effective deep learning (DL) enabled lung disease detection and classification (DL-LDDC) model with this as its goal. Using chest X-ray pictures, the proposed DLLDDC approach seeks to identify and categorize lung disease. Pre-processing, feature extraction, and classification are just a few of the processes that make up the proposed DL-LDDC approach. Pre-processing for the DL-LDDC technique is primarily divided into three phases, including contrast enhancement, thresholding-based segmentation, and region of interest (RoI) extraction using pretrained U-Net. An extensive range of simulations are performed on the benchmark chest X-ray image dataset to ensure the improved classification performance of the DL-LDDC approach. A thorough comparison of the results showed that the DL-LDDC technique outperformed the most recent state-of-the-art procedures in terms of numerous evaluation parameters.

Research on water meter reading recognition based on deep learning

At present, there are still many old-fashioned water meters in the society, and the water department needs to send staff to read the water meter after arriving at the scene with a handheld all-in-one machine. However, there are many problems in this manual meter reading method. First, a large number of meter reading work leads to low efficiency of the entire water department, consuming a lot of time and energy, and high labor costs; second, the water meters in natural scenes have problems such as serious dial contamination and other environmental factors that interfere with the meter reading staff, and the results of the meter reader cannot be verified later. In response to these problems, this paper studies a deep learning method for automatic detection and recognition of water meter readings. This paper first introduces the existing in-depth learning models, such as Faster R-CNN, SSD, and YOLOv3. Then two datasets are sorted out, one is the original water table picture dataset, and the other is a dataset cut out from the water meter image with the black bounding box showing the water meter readings. Then two plans are proposed, one is the original water table image dataset, and the other is a dataset cut out from the water meter image with the black bounding box showing the water meter readings. Finally, by comparing the three models from different angles, it is determined that YOLOv3 in the second solution has the best recognition effect, and the accuracy rate reaches 90.61%, which can greatly improve work efficiency, save labor costs, and assist auditors in reviewing the read water meter readings.

The Multilingual Picture Database

The growing interdisciplinary research field of psycholinguistics is in constant need of new and up-to-date tools which will allow researchers to answer complex questions, but also expand on languages other than English, which dominates the field. One type of such tools are picture datasets which provide naming norms for everyday objects. However, existing databases tend to be small in terms of the number of items they include, and have also been normed in a limited number of languages, despite the recent boom in multilingualism research. In this paper we present the Multilingual Picture (Multipic) database, containing naming norms and familiarity scores for 500 coloured pictures, in thirty-two languages or language varieties from around the world. The data was validated with standard methods that have been used for existing picture datasets. This is the first dataset to provide naming norms, and translation equivalents, for such a variety of languages; as such, it will be of particular value to psycholinguists and other interested researchers. The dataset has been made freely available.

Non-Destructive Defect Detection Using MASK R-CNN on Industrial Radiography of Machines

Abstract: Many industrial processes, particularly those requiring casting or welding, rely heavily on quality control. Manual quality control processes, on the other hand, are frequently time-consuming and error-prone. To address the increased demand for high-quality products, sophisticated visual inspection technologies are becoming increasingly important in manufacturing lines. Convolutional Neural Networks have recently demonstrated exceptional performance in image classification and localization tasks. Based on the Mask Region-based CNN architecture, this research proposes a solution for detecting casting errors in X-ray pictures. The suggested defect detection system conducts flaw identification and segmentation on input pictures at the same time, making it appropriate for a variety of defect detection jobs. It is demonstrated that training the network to conduct defect detection and defect instance segmentation at the same time leads in greater defect detection accuracy than training on defect detection alone. Transfer learning is used to minimizetraining data requirements while increasing the trained model's prediction accuracy. More precisely, the model is trained using two huge publically available picture datasets before being fine-tuned using a relatively modest metal casting X-ray dataset.The trained model's accuracy outperforms state-of-the-art performance on the GRIMA database of Xray images (GDXray) Castings dataset and is quick enough to be deployed in production. On the GDXray Welds dataset, the system likewise works well.A variety of in-depth research are being undertaken to investigate how transfer learning, multi-task learning, and multi-class learning affect the trained system's performance.

Traffic Sign Detection and Recognition

Street traffic signs give directions, cautioning data, to control driver conduct. Also, these signs give a dependable assurance to protected and helpful driving. The Traffic Sign Detection and Recognition (TSDR) framework is one of the essential applications for Advanced Driver Assistance Systems (ADAS). TSDR has gotten a lot of consideration over the new year’s. Be that as it may, it is as yet a difficult field of picture handling. In this work, the area of visual article discovery, following and acknowledgment in the rush hour gridlock climate is investigated. The essential spotlight is placed on the issue of video-based traffic sign acknowledgment (TSR), which is one of the significant assignments in the contemporary visual driver help frameworks. Certain calculations thusly introduced are utilized for tackling related issues, for example, walker identification or grouping of vehicle models. At the recognition stage a few procedures are broke down and assessed utilizing restrained picture and video datasets. This framework additionally investigates the chance of custom preparation of YOLOv3 based calculation.

A Fusion Algorithm of Object Detection and Tracking for Unmanned Surface Vehicles.

To provide reliable input for obstacle avoidance and decision-making, unmanned surface vehicles (USV) need to have the function of sensing the position of other USV targets in the process of cooperation and confrontation. Due to the small size of the target and the interference of the water and sky background, the current algorithms are prone to missed detection and drift problems when detecting and tracking USV. Therefore, in this paper, we propose a fusion algorithm of detection and tracking for USV targets. To solve the problem of vague features in the single-frame image, high-resolution and deep semantic information are obtained through a cross-stage partial network, and the anchor and convolution structure in the network has been improved given the characteristics of USV; besides, to meet the real-time requirements, the detected target is quickly tracked through correlation filtering, and the correlation characteristics of multi-frame images are obtained; then, the correlation characteristics are used to significantly reduce missed detection, and the tracking drift problems are corrected, combined with high-resolution semantic features of a single frame. Finally, the fusion algorithm is designed. In this paper, we constructed a picture dataset and a video dataset to test the effect of detection, tracking, and fusion algorithm separately, which proves the superiority of the fusion algorithm in this paper. The results show that, compared with a single detection algorithm and tracking algorithm, the fusion one can increase the success rate by more than 10%.

Analysis on Content Based Image Retrieval Using Image Enhancement and Deep Learning Convolutional Neural Networks

"Content-Based" means that an image contents search analyzes instead of meta data, including keywords, tags or image descriptions. The word contents could apply in this sense to colours, structures, textures or any details extracted from the picture itself. CBIR is desirable as searches relying purely on metadata depend on the quality and the completeness of annotations. CBIR method for the recovery of images from huge, unshaped image databases is commonly used. The CBIR method is used. Therefore, users are not satisfied with standard knowledge collection methods. In addition, there are more images available to users, as well as the advent of web creation and transmission networks. Consequently, there is a permanent and important output of digital images in many regions. Hence the rapid access to these enormous picture collections and the identical image retrieval from this broad image collection presents major challenges and demands efficient techniques. The efficiency of a content-based image retrieval system depends on the characteristic representation and similarity calculation. We therefore have a simple but powerful, profound, CNN-based, and feature-extraction and classification-based imaging system. Some promising results have been obtained from a range of empirical studies on a variety of CB IR tasks through the image database. Content-based image recovery systems (CBIR) allow you to find images identical to a query image among a picture dataset. The best-known CBIR system is Google's search by image feature.

DR-LL Gan: Diabetic Retinopathy Lesions Synthesis using Generative Adversarial Network

— Diabetic Retinopathy (DR) is a serious consequence of diabetes that seriously impact on the eyes and is a leading cause of blindness. If the lesions in DR arise in the central portion of the fundus, they may result in significant vision loss, which we refer to as Diabetic Macular Edema (DME). Deep learning (DL) techniques are commonly used utilized in ophthalmology for discriminative tasks such as diabetic retinopathy or age-related macular degeneration (AMD) diagnosis. Deep learning techniques typically need huge picture data sets for deep convolutional neural networks (DCNNs) training, it should be graded by human specialists. According to international protocol, it is classified into five severity categories. However, improving a grading model for high generality needs a significant quantity of balanced training data, which is challenging to obtain, especially at high levels of severity. Typical techniques for data augmentation, in many applications of deep learning in the retinal image processing domain, the difficulty of access to huge annotated datasets and legal concerns about patient privacy are limiting issues. As a result, the concept of creating synthetic retinal pictures that are indistinguishable from actual data has garnered more attention. GANs have been certain to be an effective framework for creating synthetic databases of anatomically accurate retinal fundus pictures. GANs, in particular, have garnered increasing attention in ophthalmology. in this article, we present a loss-less generative adversarial network (DR-LL GAN) to generate good resolution fundus pictures that May be adjusted to include random grading and information about the lesion. As a result, large-scale generated data may be used to train a DR grading and lesion segmentation model with more appropriate augmentation. Our model experiments evaluated on IDRID and MESSIDOR datasets, it's obtained a discrimination loss of 0.69374 and a generation loss of 1.10438, as well as a segmentation accuracy of 0.9840 in our tests. This might support in the optimization techniques of the neural network design and in computer-aided screening of medical picture, thus increasing diagnostic reliability for clinical assessment in the future of sophisticated technological healthcare.

Fault Diagnosis Technology for Ship Electrical Power System

This paper proposes a fault diagnosis method for ship electrical power systems on the basis of an improved convolutional neural network (CNN) to support normal ship operation. First, according to the mathematical model of the ship electrical power system, the simulation model of the ship electrical power system is built using the MATLAB/Simulink simulation software platform in order to understand the normal working state and fault state of the generator and load in the power system. Then, the model is simulated to generate the fault response curve, and the picture dataset of the network model is obtained. Second, a CNN fault diagnosis model is designed using TensorFlow, an open-source tool for deep learning. Finally, network model training is performed, and the optimal diagnosis results of the ship electrical power system are obtained to realize structural parameter optimization and diagnosis. The diagnosis results show that the established simulation model and improved CNN can provide support for fault diagnosis of the ship electrical power system, improve the operation stability and safety of the ship electrical power system, and ensure safety of the crew.

Underwater Image Saliency Detection Via Attention-based Mechanism

Due to complex lighting circumstances and other factors, visual saliency recognition of underwater images is not as good as visual saliency detection models that function well on land when applied to underwater environments. The dearth of large-scale annotated underwater picture datasets has further hindered the development of visual saliency detection in underwater images. we propose an attention-based approach for underwater image saliency detection to address this issue. The backbone network is pre-trained using both land and underwater image datasets, which effectively solves the problem of fewer datasets. After that, a residual refinement module incorporating a channel-space attention mechanism is merged, the model as a whole is trained end-to-end using underwater image datasets, and the model accuracy is experimentally tested. The experimental results in three underwater image datasets show that our method is better than several other methods in the mainstream evaluation indicators of the saliency detection.

Complex Texture Contour Feature Extraction of Cracks in Timber Structures of Ancient Architecture Based on YOLO Algorithm

Deep learning has achieved good results in the crack detection of roads and bridges. However, the timber structures of ancient architecture have strong orthotropic anisotropy and complex microscopic structures, and the law of cracks development is extremely complex. The image data has a large proportion of pixels, which is obviously different from the background gray value, and there is timber grain noise, thus the existing methods cannot accurately extract the complex texture contour feature of cracks. In previous studies, we have verified that YOLO v5s is effective in crack detection in timber structures of ancient architecture. However, there are many different versions of YOLO series models. In order to find a better algorithm, this paper mainly adopts three models including YOLO v3, YOLO v4s‐mish, and YOLO v5s to detect cracks in the timber structures of ancient architecture, and compares and analyzes the advantages and disadvantages of the three models. In the comparing process, we mainly have discussed the index performance of the three models in terms of training time, loss function, recall rate, and mAP value. We have summarized and analyzed the advantages and disadvantages of the three models in cracks detection of the timber structures of ancient architecture, and concluded the comparing results of the three models in cracks detection based on experiments. We published the first picture data set of cracks in timber structures of ancient architecture, and applied YOLO model in the intelligent identification field of cracks in timber structures of ancient architecture for the first time, which opened up a new idea for the intelligent operation and maintenance of the timber structures of ancient architecture.

FPGA-based Hardware Acceleration for Fruit Recognition Using SVM

Selection classification for Fruit recognition could be an absolute zone of inspection. Fruit Recognition mistreatment FPGA-based Hardware Acceleration by SVM is helpful for the observance and indexing of the fruits consistent with their kind with the peace of mind of a quick production chain. During this test, we have processed to initial replacement prime quality data-set of pictures grouped in the 5 preferred varieties of oval-shaped fruits. Honor to the fast image process techniques for the development, image resolution, quality of the algorithms leads to carry-out image process and computational tasks. In recent years, deep neural networks have a diode to the event of the many new applications associated with preciseness agriculture, as well as fruit recognition. An algorithm consumes computer power and memory, which has a significant impact on standard and performance, especially when working with large image datasets. Within the planned work, FPG is A based mostly on hardware acceleration for fruit, and recognition is mistreatment with SVM. The Support Vector Machine could be a real-time machine learning tool meant for high predicted classification accuracy through the attributes mentioned. Using SVM for embedded system programs is incredibly difficult attributable to the intensive computations needed. This will increase the attractiveness of implementing SVM on hardware platforms for reaching performance computing with the demanded value of power consumption. Finally, a difficult trade-off between meeting embedded period systems constraints and high classification accuracy has been determined.

AutoProstate: Towards Automated Reporting of Prostate MRI for Prostate Cancer Assessment Using Deep Learning.

Simple SummaryInternational guidelines recommend multiparametric magnetic resonance imaging (mpMRI) of the prostate for use by radiologists to identify lesions containing clinically significant prostate cancer, prior to confirmatory biopsy. Automatic assessment of prostate mpMRI using artificial intelligence algorithms holds a currently unrealized potential to improve the diagnostic accuracy achievable by radiologists alone, improve the reporting consistency between radiologists, and enhance reporting quality. In this work, we introduce AutoProstate: a deep learning-powered framework for automatic MRI-based prostate cancer assessment. In particular, AutoProstate utilizes patient data and biparametric MRI to populate an automatic web-based report which includes segmentations of the whole prostate, prostatic zones, and candidate clinically significant prostate cancer lesions, and in addition, several derived characteristics with clinical value are presented. Notably, AutoProstate performed well in external validation using the PICTURE study dataset, suggesting value in prospective multicentre validation, with a view towards future deployment into the prostate cancer diagnostic pathway.Multiparametric magnetic resonance imaging (mpMRI) of the prostate is used by radiologists to identify, score, and stage abnormalities that may correspond to clinically significant prostate cancer (CSPCa). Automatic assessment of prostate mpMRI using artificial intelligence algorithms may facilitate a reduction in missed cancers and unnecessary biopsies, an increase in inter-observer agreement between radiologists, and an improvement in reporting quality. In this work, we introduce AutoProstate, a deep learning-powered framework for automatic MRI-based prostate cancer assessment. AutoProstate comprises of three modules: Zone-Segmenter, CSPCa-Segmenter, and Report-Generator. Zone-Segmenter segments the prostatic zones on T2-weighted imaging, CSPCa-Segmenter detects and segments CSPCa lesions using biparametric MRI, and Report-Generator generates an automatic web-based report containing four sections: Patient Details, Prostate Size and PSA Density, Clinically Significant Lesion Candidates, and Findings Summary. In our experiment, AutoProstate was trained using the publicly available PROSTATEx dataset, and externally validated using the PICTURE dataset. Moreover, the performance of AutoProstate was compared to the performance of an experienced radiologist who prospectively read PICTURE dataset cases. In comparison to the radiologist, AutoProstate showed statistically significant improvements in prostate volume and prostate-specific antigen density estimation. Furthermore, AutoProstate matched the CSPCa lesion detection sensitivity of the radiologist, which is paramount, but produced more false positive detections.

"Automatic Object Detection on Aerial Images Using Convolutional Neural Networks"

Large quantities of aerial and satellite images are being acquired on a daily basis. Many practical applications may benefit from the analysis of such huge amounts of data. We propose an automated content-based analysis of aerial photography in this letter, which may be used to identify and label arbitrary objects or areas in high-resolution pictures. We developed a convolutional neural network-based approach for automated object identification for this purpose. In the tasks of aerial picture classification and object identification, a new two-stage method for network training is developed and validated. First, we used the UC Merced data set of aerial pictures to evaluate the suggested training method, and we were able to obtain an accuracy of about 98.6%. Second, a technique for automatically detecting objects was developed and tested. For GPGPU implementation, a processing time of approximately 30 seconds was needed for one aerial picture of size 5000 x 5000 pixels.

LinguaPix database: A megastudy of picture-naming norms

The major aim of the present megastudy of picture-naming norms was to address the shortcomings of the available picture data sets used in psychological and linguistic research by creating a new database of normed colour images that researchers from around the world can rely upon in their investigations. In order to do this, we employed a new form of normative study, namely a megastudy, whereby 1620 colour photographs of items spanning across 42 semantic categories were named and rated by a group of German speakers. This was done to establish the following linguistic norms: speech onset times (SOT), name agreement, accuracy, familiarity, visual complexity, valence, and arousal. The data, including over 64,000 audio files, were used to create the LinguaPix database of pictures, audio recordings, and linguistic norms, which to our knowledge, is the largest available research tool of its kind (http://linguapix.uni-mannheim.de). In this paper, we present the tool and the analysis of the major variables.

Computer-aided diagnosis of prostate cancer using multiparametric MRI and clinical features: A patient-level classification framework

Computer-aided diagnosis (CAD) of prostate cancer (PCa) using multiparametric magnetic resonance imaging (mpMRI) is actively being investigated as a means to provide clinical decision support to radiologists. Typically, these systems are trained using lesion annotations. However, lesion annotations are expensive to obtain and inadequate for characterizing certain tumor types e.g. diffuse tumors and MRI invisible tumors. In this work, we introduce a novel patient-level classification framework, denoted PCF, that is trained using patient-level labels only. In PCF, features are extracted from three-dimensional mpMRI and derived parameter maps using convolutional neural networks and subsequently, combined with clinical features by a multi-classifier support vector machine scheme. The output of PCF is a probability value that indicates whether a patient is harboring clinically significant PCa (Gleason score ≥3+4) or not. PCF achieved mean area under the receiver operating characteristic curves of 0.79 and 0.86 on the PICTURE and PROSTATEx datasets respectively, using five-fold cross-validation. Clinical evaluation over a temporally separated PICTURE dataset cohort demonstrated comparable sensitivity and specificity to an experienced radiologist. We envision PCF finding most utility as a second reader during routine diagnosis or as a triage tool to identify low-risk patients who do not require a clinical read.

WITHDRAWN: A quantitative segmentation and classification brain bleeding injury detection using CNN

Brain Passing on (BB) is one of the first driving causes of around the world passing and requires suitable helpful treatment. Subsequently, in this paper, our point is to classify brain computed tomography (CT) channel pictures into undesirable picture and normal picture. Our as of late proposed convolutional neural organize (CNN) appear utilizes picture combination and CNN approaches. At to begin with, a number of preprocessing operations have been utilized by utilizing multi-focus picture combination in organize to advance the quality of CT pictures. Energize, preprocessed pictures are empowered into the as of late proposed 13 layers CNN building for BB classification. The quality of our CNN procedure has been checked by conducting two tests on two different datasets. Inside the to start with attempt, CT picture dataset is allocated into 20% testing and 80% preparing sets, while inside the minute test, 10-fold cross-validation of the picture dataset has been performed. The classification accuracy gotten by our procedure on dataset 1 inside the to start with try is 98.33% and inside the minute test, it is 98.77%, while in dataset 2 accuracy gotten in test 1 and 2 is 92.22% and 93.33% independently. All the tests have been conducted on the honest to goodness CT picture dataset which we have been collected from Around the world Cancer Center (ICC), CSI remedial Mission Clinic, Neyyoor, India. The comes around gotten by the proposed procedure have besides been compared with AlexNet and ResNet50 where comes almost show up improvement over these CNN structures.