Search results for: drone images

Authors: Okike Benjamin, Garba E J. D.

Abstract:

Encryption is used to conceal information from prying eyes. Presently, information and data encryption are common due to the volume of data and information in transit across the globe on daily basis. Image encryption is yet to receive the attention of the researchers as deserved. In other words, video and multimedia documents are exposed to unauthorized accessors. The authors propose image encryption using matrix transpose. An algorithm that would allow image encryption is developed. In this proposed image encryption technique, the image to be encrypted is split into parts based on the image size. Each part is encrypted separately using matrix transpose. The actual encryption is on the picture elements (pixel) that make up the image. After encrypting each part of the image, the positions of the encrypted images are swapped before transmission of the image can take place. Swapping the positions of the images is carried out to make the encrypted image more robust for any cryptanalyst to decrypt.

Keywords: image encryption, matrices, pixel, matrix transpose

Procedia PDF Downloads 421

Authors: Rongbo Shen, Jianhua Yao, Kezhou Yan, Kuan Tian, Cheng Jiang, Ke Zhou

Abstract:

This study proposes a generative adversarial networks (GAN) framework to perform synthetic sampling in feature space, i.e., feature augmentation, to address the class imbalance problem in medical image analysis. A feature extraction network is first trained to convert images into feature space. Then the GAN framework incorporates adversarial learning to train a feature generator for the minority class through playing a minimax game with a discriminator. The feature generator then generates features for minority class from arbitrary latent distributions to balance the data between the majority class and the minority class. Additionally, a data cleaning technique, i.e., Tomek link, is employed to clean up undesirable conflicting features introduced from the feature augmentation and thus establish well-defined class clusters for the training. The experiment section evaluates the proposed method on two medical image analysis tasks, i.e., mass classification on mammogram and cancer metastasis classification on histopathological images. Experimental results suggest that the proposed method obtains superior or comparable performance over the state-of-the-art counterparts. Compared to all counterparts, our proposed method improves more than 1.5 percentage of accuracy.

Keywords: class imbalance, synthetic sampling, feature augmentation, generative adversarial networks, data cleaning

Procedia PDF Downloads 127

Authors: Rohit Semwal, Shivam Arora, Saurav, Sangita Roy

Abstract:

This work proposes a supervised framework with deep convolutional neural networks (CNNs) for vision-based sign language recognition of static gestures. Our approach addresses the acquisition and segmentation of correct inputs for the CNN-based classifier. Microsoft Kinect™ sensor, despite complex environmental conditions, can track hands efficiently. Skin Colour based segmentation is applied on cropped images of hands in different poses, used to depict different sign language gestures. The segmented hand images are used as an input for our classifier. The CNN classifier proposed in the paper is able to classify the input images with a high degree of accuracy. The system was trained and tested on 39 static sign language gestures, including 26 letters of the alphabet and 13 commonly used words. This paper includes a problem definition for building the proposed system, which acts as a sign language translator between deaf/mute and the rest of the society. It is then followed by a focus on reviewing existing knowledge in the area and work done by other researchers. It also describes the working principles behind different components of CNNs in brief. The architecture and system design specifications of the proposed system are discussed in the subsequent sections of the paper to give the reader a clear picture of the system in terms of the capability required. The design then gives the top-level details of how the proposed system meets the requirements.

Keywords: sign language, CNN, HCI, segmentation

Procedia PDF Downloads 157

Authors: Wilai Masanga, Chirapha Tannanonta, Sangutid Thongsawad, Sasikarn Chamchod, Todsaporn Fuangrod

Abstract:

The major factors of radiotherapy for head and neck (HN) cancers include patient’s anatomical changes and tumour shrinkage. These changes can significantly affect the planned dose distribution that causes the treatment plan deterioration. A measured transit EPID images compared to a predicted EPID images using gamma analysis has been clinically implemented to verify the dose accuracy as part of adaptive radiotherapy protocol. However, a global gamma analysis dose not sensitive to some critical organ changes as the entire treatment field is compared. The objective of this feasibility study is to evaluate the dosimetric response to patient anatomical changes during the treatment course in HN IMRT (Head and Neck Intensity-Modulated Radiation Therapy) using a novel comparison method; organ-of-interest gamma analysis. This method provides more sensitive to specific organ change detection. Random replanned 5 HN IMRT patients with causes of tumour shrinkage and patient weight loss that critically affect to the parotid size changes were selected and evaluated its transit dosimetry. A comprehensive physics-based model was used to generate a series of predicted transit EPID images for each gantry angle from original computed tomography (CT) and replan CT datasets. The patient structures; including left and right parotid, spinal cord, and planning target volume (PTV56) were projected to EPID level. The agreement between the transit images generated from original CT and replanned CT was quantified using gamma analysis with 3%, 3mm criteria. Moreover, only gamma pass-rate is calculated within each projected structure. The gamma pass-rate in right parotid and PTV56 between predicted transit of original CT and replan CT were 42.8%( ± 17.2%) and 54.7%( ± 21.5%). The gamma pass-rate for other projected organs were greater than 80%. Additionally, the results of organ-of-interest gamma analysis were compared with 3-dimensional cone-beam computed tomography (3D-CBCT) and the rational of replan by radiation oncologists. It showed that using only registration of 3D-CBCT to original CT does not provide the dosimetric impact of anatomical changes. Using transit EPID images with organ-of-interest gamma analysis can provide additional information for treatment plan suitability assessment.

Keywords: re-plan, anatomical change, transit electronic portal imaging device, EPID, head, and neck

Procedia PDF Downloads 216

Authors: A. Seddiki, M. Djebbouri, D. Guerchi

Abstract:

Medical imaging produces human body pictures in digital form. Since these imaging techniques produce prohibitive amounts of data, compression is necessary for storage and communication purposes. Many current compression schemes provide a very high compression rate but with considerable loss of quality. On the other hand, in some areas in medicine, it may be sufficient to maintain high image quality only in region of interest (ROI). This paper discusses a contribution to the lossless compression in the region of interest of Scintigraphic images based on SPIHT algorithm and global transform thresholding using Huffman coding.

Keywords: global thresholding transform, huffman coding, region of interest, SPIHT coding, scintigraphic images

Procedia PDF Downloads 367

Authors: Shaira L. Kee, Michael Aaron G. Sy, Myles Joshua T. Tan, Hezerul Abdul Karim, Nouar AlDahoul

Abstract:

Skin diseases are considered the fourth most common disease, with melanoma and non-melanoma skin cancer as the most common type of cancer in Caucasians. The alarming increase in Skin Cancer cases shows an urgent need for further research to improve diagnostic methods, as early diagnosis can significantly improve the 5-year survival rate. Machine Learning algorithms for image pattern analysis in diagnosing skin lesions can dramatically increase the accuracy rate of detection and decrease possible human errors. Several studies have shown the diagnostic performance of computer algorithms outperformed dermatologists. However, existing methods still need improvements to reduce diagnostic errors and generate efficient and accurate results. Our paper proposes an ensemble method to classify dermoscopic images into benign and malignant skin lesions. The experiments were conducted using the International Skin Imaging Collaboration (ISIC) image samples. The dataset contains 3,297 dermoscopic images with benign and malignant categories. The results show improvement in performance with an accuracy of 88% and an F1 score of 87%, outperforming other existing models such as support vector machine (SVM), Residual network (ResNet50), EfficientNetB0, EfficientNetB4, and VGG16.

Keywords: deep learning - VGG16 - efficientNet - CNN – ensemble – dermoscopic images - melanoma

Procedia PDF Downloads 81

Authors: U. Ravi Babu, Mohd Mastan

Abstract:

The present paper proposes a novel approach for handwritten digit recognition system. The present paper extract digit image features based on distance measure and derives an algorithm to classify the digit images. The distance measure can be performing on the thinned image. Thinning is the one of the preprocessing technique in image processing. The present paper mainly concentrated on an extraction of features from digit image for effective recognition of the numeral. To find the effectiveness of the proposed method tested on MNIST database, CENPARMI, CEDAR, and newly collected data. The proposed method is implemented on more than one lakh digit images and it gets good comparative recognition results. The percentage of the recognition is achieved about 97.32%.

Keywords: handwritten digit recognition, distance measure, MNIST database, image features

Procedia PDF Downloads 461

Authors: Raunak Munjal, Mohammad Akif Ali, Prithiv Raj

Abstract:

This study investigates the relative effectiveness of various flight controllers for drone obstacle avoidance. To assess ultrasonic sensors' performance in real-time obstacle detection, they are integrated with ESP32 and Arduino Nano controllers. The study determines which controller is most effective for this particular application by analyzing important parameters such as accuracy (mean absolute error), standard deviation, and mean distance range. Furthermore, the study explores the possibility of incorporating state-driven algorithms into the Arduino Nano configuration to potentially improve obstacle detection performance. The results offer significant perspectives for enhancing sensor integration, choosing the best flight controller for obstacle avoidance, and maybe enhancing drones' general environmental navigation ability.

Keywords: ultrasonic distance measurement, accuracy and consistency, flight controller comparisons, ESP32 vs arduino nano

Procedia PDF Downloads 58

Authors: Ping Bo, Meng Yunshan

Abstract:

Satellite-derived sea surface temperature (SST) is a key parameter for many operational and scientific applications. However, the disadvantage of SST data is a high percentage of missing data which is mainly caused by cloud coverage. Data Interpolating Empirical Orthogonal Function (DINEOF) algorithm is an EOF-based technique for reconstructing the missing data and has been widely used in oceanographic field. The reconstruction of SST images within a long time series using DINEOF can cause large discontinuities and one solution for this problem is to filter the temporal covariance matrix to reduce the spurious variability. Based on the previous researches, an algorithm is presented in this paper to improve the temporal correlations in EOF expansion. Similar with the previous researches, a filter, such as Laplacian filter, is implemented on the temporal covariance matrix, but the temporal relationship between two consecutive images which is used in the filter is considered in the presented algorithm, for example, two images in the same season are more likely correlated than those in the different seasons, hence the latter one is less weighted in the filter. The presented approach is tested for the monthly nighttime 4-km Advanced Very High Resolution Radiometer (AVHRR) Pathfinder SST for the long-term period spanning from 1989 to 2006. The results obtained from the presented algorithm are compared to those from the original DINEOF algorithm without filtering and from the DINEOF algorithm with filtering but without taking temporal relationship into account.

Keywords: data interpolating empirical orthogonal function, image reconstruction, sea surface temperature, temporal filter

Procedia PDF Downloads 324

Authors: Sedat Aktas, Egemen Ulusoy, Remzi Yildirim

Abstract:

This article explains how to get a ram image from a computer with a Linux operating system and what steps should be followed while getting it. What we mean by taking a ram image is the process of dumping the physical memory instantly and writing it to a file. This process can be likened to taking a picture of everything in the computer’s memory at that moment. This process is very important for tools that analyze ram images. Volatility can be given as an example because before these tools can analyze ram, images must be taken. These tools are used extensively in the forensic world. Forensic, on the other hand, is a set of processes for digitally examining the information on any computer or server on behalf of official authorities. In this article, the protected mode architecture in the Linux operating system is examined, and the way to save the image sample of the kernel driver and system memory to disk is followed. Tables and access methods to be used in the operating system are examined based on the basic architecture of the operating system, and the most appropriate methods and application methods are transferred to the article. Since there is no article directly related to this study on Linux in the literature, it is aimed to contribute to the literature with this study on obtaining ram images. LIME can be mentioned as a similar tool, but there is no explanation about the memory dumping method of this tool. Considering the frequency of use of these tools, the contribution of the study in the field of forensic medicine has been the main motivation of the study due to the intense studies on ram image in the field of forensics.

Keywords: linux, paging, addressing, ram-image, memory dumping, kernel modules, forensic

Procedia PDF Downloads 117

Authors: Y. Fekir, K. Mederbal, M. A. Hammadouche, D. Anteur

Abstract:

The cultural park of the Ahaggar, occupying a large area of Algeria, is characterized by a rich wetlands area to be preserved and managed both in time and space. The management of a large area, by its complexity, needs large amounts of data, which for the most part, are spatially localized (DEM, satellite images and socio-economic information...), where the use of conventional and traditional methods is quite difficult. The remote sensing, by its efficiency in environmental applications, became an indispensable solution for this kind of studies. Remote sensing imaging data have been very useful in the last decade in very interesting applications. They can aid in several domains such as the detection and identification of diverse wetland surface targets, topographical details, and geological features... In this work, we try to extract automatically wetlands area using multispectral remotely sensed data on-board the Earth Observing 1 (EO-1) and Landsat satellite. Both are high-resolution multispectral imager with a 30 m resolution. The instrument images an interesting surface area. We have used images acquired over the several area of interesting in the National Park of Ahaggar in the south of Algeria. An Extraction Algorithm is applied on the several spectral index obtained from combination of different spectral bands to extract wetlands fraction occupation of land use. The obtained results show an accuracy to distinguish wetlands area from the other lad use themes using a fine exploitation on spectral index.

Keywords: multispectral data, EO1, landsat, wetlands, Ahaggar, Algeria

Procedia PDF Downloads 377

Authors: W. R. M. Shehani Shanika

Abstract:

Selling goods and services using images, names and personalities of celebrities has become a common marketing strategy identified in modern physical and online markets. Two concepts called globalization and open economy have given numerous reasons to develop businesses to earn higher profits. Therefore, global market plus domestic markets in various countries have vigorously endorsing images of famous sport stars, film stars, singing stars and cartoon characters for the purpose of increasing demand for goods and services rendered by them. It has been evident that these trade strategies have become a threat to famous personalities in financially and personally. Right to the image is a basic human right which celebrities owned to avoid themselves from various commercial exploitations. In this respect, this paper aims to assess whether the law relating to character merchandising satisfactorily protects right to image of celebrities. However, celebrities can decide how much they receive for each representation to the general public. Simply they have exclusive right to decide monetary value for their image. But most commonly every country uses law relating to unfair competition to regulate matters arise thereof. Legal norms in unfair competition are not enough to protect image of celebrities. Therefore, celebrities must be able to avoid unauthorized use of their images for commercial purposes by fraudulent traders and getting unjustly enriched, as their images have economic value. They have the right for use their image for any commercial purpose and earn profits. Therefore it is high time to recognize right to image as a new dimension to be protected in the legal framework of character merchandising. Unfortunately, to the author’s best knowledge there are no any uniform, single international standard which recognizes right to the image of celebrities in the context of character merchandising. The paper identifies it as a controversial legal barrier faced by celebrities in the rapidly evolving marketplace. Finally, this library-based research concludes with proposals to ensure the right to image more broadly in the legal context of character merchandising.

Keywords: brand endorsement, celebrity, character merchandising, intellectual property rights, right to image, unfair competition

Procedia PDF Downloads 138

Authors: Ankit Kamboj, Shikha Talwar, Nilesh Powar

Abstract:

To successfully deliver our products in the market, the employees need to be in a safe environment, especially in an industrial and manufacturing environment. The consequences of delinquency in wearing safety glasses while working in industrial plants could be high risk to employees, hence the need to develop a real-time automatic detection system which detects the persons (violators) not wearing safety glasses. In this study a convolutional neural network (CNN) algorithm called faster region based CNN (Faster RCNN) with rotated bounding box has been used for detecting safety glasses on persons; the algorithm has an advantage of detecting safety glasses with different orientation angles on the persons. The proposed method of rotational bounding boxes with a convolutional neural network first detects a person from the images, and then the method detects whether the person is wearing safety glasses or not. The video data is captured at the entrance of restricted zones of the industrial environment (manufacturing plant), which is further converted into images at 2 frames per second. In the first step, the CNN with pre-trained weights on COCO dataset is used for person detection where the detections are cropped as images. Then the safety goggles are labelled on the cropped images using the image labelling tool called roLabelImg, which is used to annotate the ground truth values of rotated objects more accurately, and the annotations obtained are further modified to depict four coordinates of the rectangular bounding box. Next, the faster RCNN with rotated bounding box is used to detect safety goggles, which is then compared with traditional bounding box faster RCNN in terms of detection accuracy (average precision), which shows the effectiveness of the proposed method for detection of rotatory objects. The deep learning benchmarking is done on a Dell workstation with a 16GB Nvidia GPU.

Keywords: CNN, deep learning, faster RCNN, roLabelImg rotated bounding box, safety goggle detection

Procedia PDF Downloads 130

Authors: Jonathan Gong

Abstract:

Purpose AI-driven solutions are at the forefront of many pathology and medical imaging methods. Using algorithms designed to better the experience of medical professionals within their respective fields, the efficiency and accuracy of diagnosis can improve. In particular, X-rays are a fast and relatively inexpensive test that can diagnose diseases. In recent years, X-rays have not been widely used to detect and diagnose COVID-19. The under use of Xrays is mainly due to the low diagnostic accuracy and confounding with pneumonia, another respiratory disease. However, research in this field has expressed a possibility that artificial neural networks can successfully diagnose COVID-19 with high accuracy. Models and Data The dataset used is the COVID-19 Radiography Database. This dataset includes images and masks of chest X-rays under the labels of COVID-19, normal, and pneumonia. The classification model developed uses an autoencoder and a pre-trained convolutional neural network (DenseNet201) to provide transfer learning to the model. The model then uses a deep neural network to finalize the feature extraction and predict the diagnosis for the input image. This model was trained on 4035 images and validated on 807 separate images from the ones used for training. The images used to train the classification model include an important feature: the pictures are cropped beforehand to eliminate distractions when training the model. The image segmentation model uses an improved U-Net architecture. This model is used to extract the lung mask from the chest X-ray image. The model is trained on 8577 images and validated on a validation split of 20%. These models are calculated using the external dataset for validation. The models’ accuracy, precision, recall, f1-score, IOU, and loss are calculated. Results The classification model achieved an accuracy of 97.65% and a loss of 0.1234 when differentiating COVID19-infected, pneumonia-infected, and normal lung X-rays. The segmentation model achieved an accuracy of 97.31% and an IOU of 0.928. Conclusion The models proposed can detect COVID-19, pneumonia, and normal lungs with high accuracy and derive the lung mask from a chest X-ray with similarly high accuracy. The hope is for these models to elevate the experience of medical professionals and provide insight into the future of the methods used.

Keywords: artificial intelligence, convolutional neural networks, deep learning, image processing, machine learning

Procedia PDF Downloads 130

Authors: Anshika Kankane, Dongshik Kang

Abstract:

Marine debris has a significant influence on coastal environments, damaging biodiversity, and causing loss and damage to marine and ocean sector. A functional cost-effective and automatic approach has been used to look up at this problem. Computer vision combined with a deep learning-based model is being proposed to identify and categorize marine debris of seven kinds on different beach locations of Japan. This research compares state-of-the-art deep learning models with a suggested model architecture that is utilized as a feature extractor for debris categorization. The model is being proposed to detect seven categories of litter using a manually constructed debris dataset, with the help of Mask R-CNN for instance segmentation and a shape matching network called HOGShape, which can then be cleaned on time by clean-up organizations using warning notifications of the system. The manually constructed dataset for this system is created by annotating the images taken by fixed KaKaXi camera using CVAT annotation tool with seven kinds of category labels. A pre-trained HOG feature extractor on LIBSVM is being used along with multiple templates matching on HOG maps of images and HOG maps of templates to improve the predicted masked images obtained via Mask R-CNN training. This system intends to timely alert the cleanup organizations with the warning notifications using live recorded beach debris data. The suggested network results in the improvement of misclassified debris masks of debris objects with different illuminations, shapes, viewpoints and litter with occlusions which have vague visibility.

Keywords: computer vision, debris, deep learning, fixed live camera images, histogram of gradients feature extractor, instance segmentation, manually annotated dataset, multiple template matching

Procedia PDF Downloads 106

Authors: Hana Rabbouch

Abstract:

In recent years, there has been considerable growth of denoising techniques mainly devoted to medical imaging. This important evolution is not only due to the progress of computing techniques, but also to the emergence of multi-resolution analysis (MRA) on both mathematical and algorithmic bases. In this paper, a comparative study is conducted between the two best-known MRA-based decomposition techniques: the Empirical Mode Decomposition (EMD) and the Discrete Wavelet Transform (DWT). The comparison is carried out in a framework of multi-scale denoising, where a Non-Local Means (NLM) filter is performed scale-by-scale to a sample of benchmark medical images. The results prove the effectiveness of the multiscaled denoising, especially when the NLM filtering is coupled with the EMD.

Keywords: medical imaging, non local means, denoising, multiscaled analysis, empirical mode decomposition, wavelets

Procedia PDF Downloads 141

Authors: Nadia Ben Youssef, Aicha Bouzid

Abstract:

Gradient estimation is one of the most fundamental tasks in the field of image processing in general, and more particularly for color images since that the research in color image gradient remains limited. The widely used gradient method is Di Zenzo’s gradient operator, which is based on the measure of squared local contrast of color images. The proposed gradient mechanism, presented in this paper, is based on the principle of the Di Zenzo’s approach using quaternion representation. This edge detector is compared to a marginal approach based on multiscale product of wavelet transform and another vector approach based on quaternion convolution and vector gradient approach. The experimental results indicate that the proposed color gradient operator outperforms marginal approach, however, it is less efficient then the second vector approach.

Keywords: gradient, edge detection, color image, quaternion

Procedia PDF Downloads 234

Authors: Jean-Francois Rajotte, Martin Sotir, Frank Gouineau

Abstract:

The installation and management of traffic monitoring devices can be costly from both a financial and resource point of view. It is therefore important to take advantage of in-place infrastructures to extract the most information. Here we show how low-quality urban road traffic images from cameras already available in many cities (such as Montreal, Vancouver, and Toronto) can be used to estimate traffic flow. To this end, we use a pre-trained neural network, developed for object detection, to count vehicles within images. We then compare the results with human annotations gathered through crowdsourcing campaigns. We use this comparison to assess performance and calibrate the neural network annotations. As a use case, we consider six months of continuous monitoring over hundreds of cameras installed in the city of Montreal. We compare the results with city-provided manual traffic counting performed in similar conditions at the same location. The good performance of our system allows us to consider applications which can monitor the traffic conditions in near real-time, making the counting usable for traffic-related services. Furthermore, the resulting annotations pave the way for building a historical vehicle counting dataset to be used for analysing the impact of road traffic on many city-related issues, such as urban planning, security, and pollution.

Keywords: traffic monitoring, deep learning, image annotation, vehicles, roads, artificial intelligence, real-time systems

Procedia PDF Downloads 200

Authors: Zeyu Zhou, Xiaojun Bi

Abstract:

Due to the limited environmental parameters and the limited resolution of the sensor, the universal existence of the mixed pixels in the process of remote sensing images restricts the spatial resolution of the remote sensing images. Sub-pixel mapping technology can effectively improve the spatial resolution. As the bilinear interpolation algorithm inevitably produces the edge blur effect, which leads to the inaccurate sub-pixel mapping results. In order to avoid the edge blur effect that affects the sub-pixel mapping results in the interpolation process, this paper presents a new edge-directed interpolation algorithm which uses the covariance adaptive interpolation algorithm on the edge of the low-resolution image and uses bilinear interpolation algorithm in the low-resolution image smooth area. By using the edge-directed interpolation algorithm, the super-resolution of the image with low resolution is obtained, and we get the percentage of each sub-pixel under a certain type of high-resolution image. Then we rely on the probability value as a soft attribute estimate and carry out sub-pixel scale under the ‘hard classification’. Finally, we get the result of sub-pixel mapping. Through the experiment, we compare the algorithm and the bilinear algorithm given in this paper to the results of the sub-pixel mapping method. It is found that the sub-pixel mapping method based on the edge-directed interpolation algorithm has better edge effect and higher mapping accuracy. The results of the paper meet our original intention of the question. At the same time, the method does not require iterative computation and training of samples, making it easier to implement.

Keywords: remote sensing images, sub-pixel mapping, bilinear interpolation, edge-directed interpolation

Procedia PDF Downloads 229

Authors: Adrien Marque, Daniel Delahaye, Pierre Maréchal, Isabelle Berry

Abstract:

Wind direction and uncertainty are crucial in aircraft or unmanned aerial vehicle trajectories. By computing wind covariance matrices on each spatial grid point, these spatial grids can be defined as images with symmetric positive definite matrix elements. A data pre-processing step, a specific convolution, a specific max-pooling, and a specific flatten layers are implemented to process such images. Then, the neural network is applied to spatial grids, whose elements are wind covariance matrices, to solve classification problems related to the feasibility of unmanned aerial vehicles based on wind direction and wind uncertainty.

Keywords: wind direction, uncertainty level, unmanned aerial vehicle, convolution neural network, SPD matrices

Procedia PDF Downloads 49

Authors: Daoudi Abdelaziz, Mahmoudi Saïd, Chikh Mohamed Amine

Abstract:

This paper presents a fully automatic segmentation method of the left ventricle at End Systolic (ES) and End Diastolic (ED) in the ultrasound images by means of an implicit deformable model (level set) based on Geodesic Active Contour model. A pre-processing Gaussian smoothing stage is applied to the image, which is essential for a good segmentation. Before the segmentation phase, we locate automatically the area of the left ventricle by using a detection approach based on the Hough Transform method. Consequently, the result obtained is used to automate the initialization of the level set model. This initial curve (zero level set) deforms to search the Endocardial border in the image. On the other hand, quantitative evaluation was performed on a data set composed of 15 subjects with a comparison to ground truth (manual segmentation).

Keywords: level set method, transform Hough, Gaussian smoothing, left ventricle, ultrasound images.

Procedia PDF Downloads 465

Authors: Abhisek Dash

Abstract:

This study had three neural networks, one for number segmentation, one for number detection and one for number recognition all of which are coupled to one another. All networks were trained on the MNIST dataset and were convolutional. It was assumed that the images had lighter background and darker foreground. The segmentation network took 28x28 images as input and had sixteen outputs. Segmentation training starts when a dark pixel is encountered. Taking a window(7x7) over that pixel as focus, the eight neighborhood of the focus was checked for further dark pixels. The segmentation network was then trained to move in those directions which had dark pixels. To this end the segmentation network had 16 outputs. They were arranged as “go east”, ”don’t go east ”, “go south east”, “don’t go south east”, “go south”, “don’t go south” and so on w.r.t focus window. The focus window was resized into a 28x28 image and the network was trained to consider those neighborhoods which had dark pixels. The neighborhoods which had dark pixels were pushed into a queue in a particular order. The neighborhoods were then popped one at a time stitched to the existing partial image of the number one at a time and trained on which neighborhoods to consider when the new partial image was presented. The above process was repeated until the image was fully covered by the 7x7 neighborhoods and there were no more uncovered black pixels. During testing the network scans and looks for the first dark pixel. From here on the network predicts which neighborhoods to consider and segments the image. After this step the group of neighborhoods are passed into the detection network. The detection network took 28x28 images as input and had two outputs denoting whether a number was detected or not. Since the ground truth of the bounds of a number was known during training the detection network outputted in favor of number not found until the bounds were not met and vice versa. The recognition network was a standard CNN that also took 28x28 images and had 10 outputs for recognition of numbers from 0 to 9. This network was activated only when the detection network votes in favor of number detected. The above methodology could segment connected and overlapping numbers. Additionally the recognition unit was only invoked when a number was detected which minimized false positives. It also eliminated the need for rules of thumb as segmentation is learned. The strategy can also be extended to other characters as well.

Keywords: convolutional neural networks, OCR, text detection, text segmentation

Procedia PDF Downloads 161

Authors: Chang-Hsing Lee, Cheng-Chang Lien, Chin-Chuan Han

Abstract:

In this paper, an edge-strength guided multiscale retinex (EGMSR) approach will be proposed for color image contrast enhancement. In EGMSR, the pixel-dependent weight associated with each pixel in the single scale retinex output image is computed according to the edge strength around this pixel in order to prevent from over-enhancing the noises contained in the smooth dark/bright regions. Further, by fusing together the enhanced results of EGMSR and adaptive multiscale retinex (AMSR), we can get a natural fused image having high contrast and proper tonal rendition. Experimental results on several low-contrast images have shown that our proposed approach can produce natural and appealing enhanced images.

Keywords: image enhancement, multiscale retinex, image fusion, EGMSR

Procedia PDF Downloads 458

Authors: Thibaud Berthomier, Ali Mansour, Luc Bressollette, Frédéric Le Roy, Dominique Mottier, Léo Fréchier, Barthélémy Hermenault

Abstract:

Deep Venous Thrombosis (DVT) occurs when a thrombus is formed within a deep vein (most often in the legs). This disease can be deadly if a part or the whole thrombus reaches the lung and causes a Pulmonary Embolism (PE). This disorder, often asymptomatic, has multifactorial causes: immobilization, surgery, pregnancy, age, cancers, and genetic variations. Our project aims to relate the thrombus epidemiology (origins, patient predispositions, PE) to its structure using ultrasound images. Ultrasonography and elastography were collected using Toshiba Aplio 500 at Brest Hospital. This manuscript compares two classification approaches: spectral clustering and scattering operator. The former is based on the graph and matrix theories while the latter cascades wavelet convolutions with nonlinear modulus and averaging operators.

Keywords: deep venous thrombosis, ultrasonography, elastography, scattering operator, wavelet, spectral clustering

Procedia PDF Downloads 479

Authors: A. M. John-Otumu, M. M. Rahman, M. C. Onuoha, E. P. Ojonugwa

Abstract:

A Convolutional Neural Network (CNN) model for predicting Lassa fever was built using Python 3.8.0 programming language, alongside Keras 2.2.4 and TensorFlow 2.6.1 libraries as the development environment in order to reduce the current high risk of Lassa fever in West Africa, particularly in Nigeria. The study was prompted by some major flaws in existing conventional laboratory equipment for diagnosing Lassa fever (RT-PCR), as well as flaws in AI-based techniques that have been used for probing and prognosis of Lassa fever based on literature. There were 15,679 blood smear microscopic image datasets collected in total. The proposed model was trained on 70% of the dataset and tested on 30% of the microscopic images in avoid overfitting. A 3x3x3 convolution filter was also used in the proposed system to extract features from microscopic images. The proposed CNN-based model had a recall value of 96%, a precision value of 93%, an F1 score of 95%, and an accuracy of 94% in predicting and accurately classifying the images into clean or infected samples. Based on empirical evidence from the results of the literature consulted, the proposed model outperformed other existing AI-based techniques evaluated. If properly deployed, the model will assist physicians, medical laboratory scientists, and patients in making accurate diagnoses for Lassa fever cases, allowing the mortality rate due to the Lassa fever virus to be reduced through sound decision-making.

Keywords: artificial intelligence, ANN, blood smear, CNN, deep learning, Lassa fever

Procedia PDF Downloads 120

Authors: Lae-Jeong Park, Kyusoo Chung, Jungho Moon

Abstract:

This paper introduces an effective method of segmenting Korean text (place names in Korean) from a Korean road sign image. A Korean advanced directional road sign is composed of several types of visual information such as arrows, place names in Korean and English, and route numbers. Automatic classification of the visual information and extraction of Korean place names from the road sign images make it possible to avoid a lot of manual inputs to a database system for management of road signs nationwide. We propose a series of problem-specific heuristics that correctly segments Korean place names, which is the most crucial information, from the other information by leaving out non-text information effectively. The experimental results with a dataset of 368 road sign images show 96% of the detection rate per Korean place name and 84% per road sign image.

Keywords: segmentation, road signs, characters, classification

Procedia PDF Downloads 444

Authors: Dionisio Andujar, Juan lópez-Correa, Hugo Moreno, Angela Ri

Abstract:

The increased number of weeds in tomato crops highly lower yields. Weed identification with the aim of machine learning is important to carry out site-specific control. The last advances in computer vision are a powerful tool to face the problem. The analysis of RGB (Red, Green, Blue) images through Artificial Neural Networks had been rapidly developed in the past few years, providing new methods for weed classification. The development of the algorithms for crop and weed species classification looks for a real-time classification system using Object Detection algorithms based on Convolutional Neural Networks. The site study was located in commercial corn fields. The classification system has been tested. The procedure can detect and classify weed seedlings in tomato fields. The input to the Neural Network was a set of 10,000 RGB images with a natural infestation of Cyperus rotundus l., Echinochloa crus galli L., Setaria italica L., Portulaca oeracea L., and Solanum nigrum L. The validation process was done with a random selection of RGB images containing the aforementioned species. The mean average precision (mAP) was established as the metric for object detection. The results showed agreements higher than 95 %. The system will provide the input for an online spraying system. Thus, this work plays an important role in Site Specific Weed Management by reducing herbicide use in a single step.

Keywords: deep learning, object detection, cnn, tomato, weeds

Procedia PDF Downloads 103

Authors: Rauf R. Hussein, Devon M. Ramey

Abstract:

Integrating Ground Penetrating Radar (GPR) with Geographic Information Systems (GIS) can provide valuable insights for various applications, such as archaeology, transportation, and utility locating. Although there has been progress toward automating the integration of GPR data with GIS, fully automatic integration has not been achieved yet. Additionally, manually integrating GPR data with GIS can be a time-consuming and error-prone process. In this study, actual, real-world GPR applications are presented, and a software named GPR-GIS 10 is created to interactively extract subsurface targets from GPR radargrams and automatically integrate them into GIS. With this software, it is possible to quickly and reliably integrate the two techniques to create informative subsurface maps. The results indicated that automatic integration of GPR with GIS can be an efficient tool to map and view any subsurface targets in their appropriate location in a 3D space with the needed precision. The findings of this study could help GPR-GIS integrators save time and reduce errors in many GPR-GIS applications.

Keywords: GPR, GIS, GPR-GIS 10, drone technology, automation

Procedia PDF Downloads 92

Authors: Omar Elharrouss, Youssef Hmamouche, Assia Kamal Idrissi, Btissam El Khamlichi, Amal El Fallah-Seghrouchni

Abstract:

Edge detection is represented as one of the most challenging tasks in computer vision, due to the complexity of detecting the edges or boundaries in real-world images that contains objects of different types and scales like trees, building as well as various backgrounds. Edge detection is represented also as a key task for many computer vision applications. Using a set of backbones as well as attention modules, deep-learning-based methods improved the detection of edges compared with the traditional methods like Sobel and Canny. However, images of complex scenes still represent a challenge for these methods. Also, the detected edges using the existing approaches suffer from non-refined results while the image output contains many erroneous edges. To overcome this, n this paper, by using the mechanism of residual learning, a refined edge detection network is proposed (RED-Net). By maintaining the high resolution of edges during the training process, and conserving the resolution of the edge image during the network stage, we make the pooling outputs at each stage connected with the output of the previous layer. Also, after each layer, we use an affined batch normalization layer as an erosion operation for the homogeneous region in the image. The proposed methods are evaluated using the most challenging datasets including BSDS500, NYUD, and Multicue. The obtained results outperform the designed edge detection networks in terms of performance metrics and quality of output images.

Keywords: edge detection, convolutional neural networks, deep learning, scale-representation, backbone

Procedia PDF Downloads 102

Authors: Sorin Valcan, Mihail Gaianu

Abstract:

Labeling is a very costly and time consuming process which aims to generate datasets for training neural networks in several functionalities and projects. For driver monitoring system projects, the need for labeled images has a significant impact on the budget and distribution of effort. This paper presents the modifications done to an algorithm used for the generation of ground truth data for 2D eyes location on infrared images with drivers in order to improve the quality of the data and performance of the trained neural networks. The algorithm restrictions become tougher, which makes it more accurate but also less constant. The resulting dataset becomes smaller and shall not be altered by any kind of manual label adjustment before being used in the neural networks training process. These changes resulted in a much better performance of the trained neural networks.

Keywords: labeling automation, infrared camera, driver monitoring, eye detection, convolutional neural networks

Procedia PDF Downloads 117