Engineering Commons^™

Image Filtering To Improve Maize Tassel Detection Accuracy Using Machine Learning Algorithms, Eric Rodene, Gayara Demini Fernando, Ved Piyush, Yufeng Ge, James C. Schnable, Souparno Ghosh, Jinliang Yang

Department of Biological Systems Engineering: Papers and Publications

Unmanned aerial vehicle (UAV)-based imagery has become widely used to collect timeseries agronomic data, which are then incorporated into plant breeding programs to enhance crop improvements. To make efficient analysis possible, in this study, by leveraging an aerial photography dataset for a field trial of 233 different inbred lines from the maize diversity panel, we developed machine learning methods for obtaining automated tassel counts at the plot level. We employed both an object-based counting-by-detection (CBD) approach and a density-based counting-by-regression (CBR) approach. Using an image segmentation method that removes most of the pixels not associated with the plant tassels, the …

Pathology Slide Segmentation, Mohamed Mohamed

Computer Science and Engineering Theses

The goal of this project was to create an image segmentation model that would extract an image of H&E pathology slides from real life scenarios. This was a part of a project that required the extraction of an image, querying of the image, and displaying the results. The baselines of the model were to be efficient, run on mobile devices, and be able to work with cameras of varying resolutions.

Wavelet-Based Harmonization Of Local And Global Model Shifts In Federated Learning For Histopathological Images, W. Farzana, A. Temtam, K. M. Iftekharuddin

Electrical & Computer Engineering Faculty Publications

Federated Learning (FL) is a promising machine learning approach for development of data-driven global model using collaborative local models across multiple local institutions. However, the heterogeneity of medical imaging data is one of the challenges within FL. This heterogeneity is caused by the variation in imaging scanner protocols across institutions, which may result in weight shift among local models leading to deterioration in predictive accuracy of global model. The prevailing approaches involve applying different FL averaging techniques to enhance the performance of the global model, ignoring the distinct imaging features of the local domain. In this work, we address both …

A Survey On Few-Shot Class-Incremental Learning, Songsong Tian, Lusi Li, Weijun Li, Hang Ran, Xin Ning, Prayag Tiwari

Computer Science Faculty Publications

Large deep learning models are impressive, but they struggle when real-time data is not available. Few-shot class-incremental learning (FSCIL) poses a significant challenge for deep neural networks to learn new tasks from just a few labeled samples without forgetting the previously learned ones. This setup can easily leads to catastrophic forgetting and overfitting problems, severely affecting model performance. Studying FSCIL helps overcome deep learning model limitations on data volume and acquisition time, while improving practicality and adaptability of machine learning models. This paper provides a comprehensive survey on FSCIL. Unlike previous surveys, we aim to synthesize few-shot learning and incremental …

Image Segmentation With Human-In-The-Loop In Automated De-Caking Process For Powder Bed Additive Manufacturing, Vincent Opare Addo Asare-Manu

Theses and Dissertations

Additive manufacturing (AM) becomes a critical technology that increases the speed and flexibility of production and reduces the lead time for high-mix, low-volume manufacturing. One of the major bottlenecks in further increasing its productivity lies around its post-processing procedures. This work focuses on tackling a critical and inevitable step in powder-bed additive manufacturing processes, i.e., powder cleaning or de-caking. Pressing concerns can be raised with human involvement when performing this task manually. Therefore, a robot-driven automatic powder cleaning system could be an alternative to reducing time consumption and increasing safety for AM operators. However, since the color and surface texture …

Skin Lesion Segmentation In Dermoscopic Images With Noisy Data, Norsang Lama, Jason Hagerty, Anand Nambisan, Ronald Joe Stanley, William Van Stoecker

Electrical and Computer Engineering Faculty Research & Creative Works

We Propose a Deep Learning Approach to Segment the Skin Lesion in Dermoscopic Images. the Proposed Network Architecture Uses a Pretrained Efficient Net Model in the Encoder and Squeeze-And-Excitation Residual Structures in the Decoder. We Applied This Approach on the Publicly Available International Skin Imaging Collaboration (ISIC) 2017 Challenge Skin Lesion Segmentation Dataset. This Benchmark Dataset Has Been Widely Used in Previous Studies. We Observed Many Inaccurate or Noisy Ground Truth Labels. to Reduce Noisy Data, We Manually Sorted All Ground Truth Labels into Three Categories — Good, Mildly Noisy, and Noisy Labels. Furthermore, We Investigated the Effect of Such …

Detection Of Various Dental Conditions On Dental Panoramic Radiography Using Faster R-Cnn, Shih Lun Chen, Tsung Yi Chen, Yi Cheng Mao, Szu Yin Lin, Ya Yun Huang, Chiung An Chen, Yuan Jin Lin, Mian Heng Chuang, Patricia Angela R. Abu

Department of Information Systems & Computer Science Faculty Publications

The dental panoramic radiograph (DPR) is a pivotal diagnostic tool in dentistry. However, despite the growing prevalence of artificial intelligence (AI) across various medical domains, manual methods remain the prevailing means of interpreting DPR images. This study aims to introduce an advanced identification system for detecting seven dental conditions in DPR images by utilizing Faster R-CNN. The primary objectives are to enhance dentists' efficiency and evaluate the performance of various CNN models as foundational training networks. This study contributes significantly to the field in several notable ways. Firstly, including a Butterworth filter in the training process yielded an approximately 7% …

Analyzing The Benthic Cover Of Crustose Coralline Algae Using Mask-R Cnn, Rachana Ravindra

Master's Projects

Coral reefs, supporting 25% of marine biodiversity, confront challenges from local and global impacts like overfishing, runoff, acidification, and warming. Crustose Coralline Algae (CCA), pivotal for reef structure and coral settlement, are underrepresented in research. Current methods like Coral Point Count with Excel Extensions (CPCe) have limitations, relying on image quality and being time-consuming. This paper proposes computer vision and Mask R-CNN, a supervised machine learning model, for CCA analysis in reef images, considering color, texture, and shape. Results indicate promise in clustering and classifying organisms. The innovative technology reduces manual labor, enhancing image analysis, simplifying the understanding of CCA’s …

Automatic Optical Inspection-Based Pcb Fault Detection Using Image Processing, Shruti Rajiv Vaidya

Dissertations, Master's Theses and Master's Reports

Increased Printed Circuit Board (PCB) route complexity and density combined with the growing demand for low-scale rapid prototyping has increased the desire for Automated Optical Inspection (AOI) that reduces prototyping time and production costs by detecting defects early in the production process. Traditional defect detection method of human visual inspection is not only error prone but is also time-consuming given the growing complex and dense circuitry of modern-day electronics. Electric contact-based testing, either in the form of a bed of nails testing fixture or a flying probe system, is costly for low-rate rapid prototyping. An AOI is a non-contact test …

Class Activation Mapping And Uncertainty Estimation In Multi-Organ Segmentation, Md. Shibly Sadique, Walia Farzana, Ahmed Temtam, Khan Iftekharuddin, Khan Iftekharuddin (Ed.), Weijie Chen (Ed.)

Electrical & Computer Engineering Faculty Publications

Deep learning (DL)-based medical imaging and image segmentation algorithms achieve impressive performance on many benchmarks. Yet the efficacy of deep learning methods for future clinical applications may become questionable due to the lack of ability to reason with uncertainty and interpret probable areas of failures in prediction decisions. Therefore, it is desired that such a deep learning model for segmentation classification is able to reliably predict its confidence measure and map back to the original imaging cases to interpret the prediction decisions. In this work, uncertainty estimation for multiorgan segmentation task is evaluated to interpret the predictive modeling in DL …

An Exploration Of Recent Intelligent Image Analysis Techniques For Visual Pavement Surface Condition Assessment., Waqar Shahid Qureshi, Ibrahim Hassan Syed, Susan Mckeever, David Power, Brian Mulry, Kieran Feighan, Dympna O'Sullivan

Articles

Road pavement condition assessment is essential for maintenance, asset management, and budgeting for pavement infrastructure. Countries allocate a substantial annual budget to maintain and improve local, regional, and national highways. Pavement condition is assessed by measuring several pavement characteristics such as roughness, surface skid resistance, pavement strength, deflection, and visual surface distresses. Visual inspection identifies and quantifies surface distresses, and the condition is assessed using standard rating scales. This paper critically analyzes the research trends in the academic literature, professional practices and current commercial solutions for surface condition ratings by civil authorities. We observe that various surface condition rating systems …

An Introductory Module In Medical Image Segmentation For Bme Students, Christine Buffinton, Donna Ebenstein, James W. Baish

Faculty Journal Articles

To support recent trends toward the use of patient-specific anatomical models from medical imaging data, we present a learning module for use in the undergraduate BME curriculum that introduces image segmentation, the process of partitioning digital images to isolate specific anatomical features. Five commercially available software packages were evaluated based on their perceived learning curve, ease of use, tools for segmentation and rendering, special tools, and cost: ITK-SNAP, 3D Slicer, OsiriX, Mimics, and Amira. After selecting the package best suited for a stand-alone course module on medical image segmentation, instructional materials were developed that included a general introduction to imaging, …

A Deep Learning Approach For Airport Runway Identification From Satellite Imagery, Mahmut Gemici

Theses and Dissertations

The United States lacks a comprehensive national database of private Prior Permission Required (PPR) airports. The primary reason such a database does not exist is that there are no federal regulatory obligations for these facilities to have their information re-evaluated or updated by the Federal Aviation Administration (FAA) or the local state Department of Transportation (DOT) once the data has been entered into the system. The often outdated and incorrect information about landing sites presents a serious risk factor in aviation safety. In this thesis, we present a machine learning approach for detecting airport landing sites from Google Earth satellite …

A Deep Learning Image Segmentation Model For Agricultural Irrigation System Classification, Ehsan Raei, Ata Akbari Asanjan, Mohammad Reza Nikoo, Mojtaba Sadegh, Shokoufeh Pourshahabi, Jan Franklin Adamowski

Civil Engineering Faculty Publications and Presentations

Effective water management requires a large-scale understanding of agricultural irrigation systems and how they shift in response to various stressors. Here, we leveraged advances in Machine Learning and availability of very high resolution remote sensing imagery to help resolve this long-standing issue. To this end, we developed a deep learning model to classify irrigation systems at a regional scale using remote sensing imagery. After testing different model architectures, hyper parameters, class weights and image sizes, we selected a U-Net architecture with a Resnet-34 backbone for this purpose. We applied transfer learning to increase training efficiency and model performance. We considered …

Research On Visual Inspection Algorithm Of Crimping Appearance Defects For Wiring Harness Terminals, Bingan Yuan, Mingen Zhong, Jingxin Ni

Journal of System Simulation

Abstract: Aiming at the low efficiency and high missing rate of wiring harness terminals, an image detection method based on machine vision is proposed. The characteristic parameters of five typical defects in three main parts of wiring harness terminals are analyzed and defined. Tthe algorithms of extracting positioning datum, segmenting inspected-parts adaptively, extracting the defect features and calculating the characteristic parameters are designed respectively, and the defects criterions are given. The experimental results show that the algorithms are suitable for single defect and multi-class defects, both the miss detection rate and the false positiveness rate are low. The accuracy and …

Image And Video Segmentation Of Appearance-Volatile Objects, Yongqing Liang

LSU Master's Theses

Segmentation is a process of partitioning a digital image or frame into multiple regions or objects. The goal of segmentation is to identify and locate the objects of interest with their boundaries. Recent segmentation approaches often follow such a pipeline: they first train the model on a collected dataset and then evaluate the trained model on a given image or video. They assume that the appearance of object is consistent in training and testing sets. However, the appearance of object may change in different photography conditions. How to effectively segment the objects with volatile appearance remains under-explored. In this work, …

Monocular Semantic Slam Method Based On Object Relation Description, Shiqi Lin, Jikai Wang, Haoyuan Pei, Hao Zhao, Zonghai Chen

Journal of System Simulation

Abstract: Semantic information perception of the external environment and accurate positioning are the keys to autonomous navigation and operation of mobile robots. This paper proposes a method of semantic simultaneous localization and mapping (SLAM) based on a monocular camera. The system completes three-dimensional (3D) object detection while estimating the trajectory. We model the 3D objects with cuboids. Then, the semantic meanings, color distribution, size and neighborhood topology of the objects are extracted as descriptors for the accurate matching of objects between different frames. The camera pose, map points and object landmarks are optimized jointly in the backend of the system. …

Is Contrastive Learning Suitable For Left Ventricular Segmentation In Echocardiographic Images?, Mohamed Saeed, Rand Muhtaseb, Mohammad Yaqub

Computer Vision Faculty Publications

Contrastive learning has proven useful in many applications where access to labelled data is limited. The lack of annotated data is particularly problematic in medical image segmenta-tion as it is difficult to have clinical experts manually annotate large volumes of data. One such task is the segmentation of cardiac structures in ultrasound images of the heart. In this paper, we argue whether or not contrastive pretraining is helpful for the segmentation of the left ventricle in echocardiography images. Furthermore, we study the effect of this on two segmentation networks, DeepLabV3, as well as the commonly used segmentation net-work, UNet. Our …

Chimeranet: U-Net For Hair Detection In Dermoscopic Skin Lesion Images, Norsang Lama, Reda Kasmi, Jason R. Hagerty, R. Joe Stanley, Reagan Harris Young, Jessica Miinch, Januka Nepal, Anand Nambisan, William V. Stoecker

Electrical and Computer Engineering Faculty Research & Creative Works

Hair and ruler mark structures in dermoscopic images are an obstacle preventing accurate image segmentation and detection of critical network features. Recognition and removal of hairs from images can be challenging, especially for hairs that are thin, overlapping, faded, or of similar color as skin or overlaid on a textured lesion. This paper proposes a novel deep learning (DL) technique to detect hair and ruler marks in skin lesion images. Our proposed ChimeraNet is an encoder-decoder architecture that employs pretrained EfficientNet in the encoder and squeeze-and-excitation residual (SERes) structures in the decoder. We applied this approach at multiple image sizes …

A Quantitative Validation Of Multi-Modal Image Fusion And Segmentation For Object Detection And Tracking, Nicholas Lahaye, Michael J. Garay, Brian D. Bue, Hesham El-Askary, Erik Linstead

Mathematics, Physics, and Computer Science Faculty Articles and Research

In previous works, we have shown the efficacy of using Deep Belief Networks, paired with clustering, to identify distinct classes of objects within remotely sensed data via cluster analysis and qualitative analysis of the output data in comparison with reference data. In this paper, we quantitatively validate the methodology against datasets currently being generated and used within the remote sensing community, as well as show the capabilities and benefits of the data fusion methodologies used. The experiments run take the output of our unsupervised fusion and segmentation methodology and map them to various labeled datasets at different levels of global …

Machine Learning-Based Recognition On Crowdsourced Food Images, Aditya Kulkarni

Honors Scholar Theses

With nearly a third of the world’s population suffering from food-induced chronic diseases such as obesity, the role of food in community health is required now more than ever. While current research underscores food proximity and density, there is a dearth in regard to its nutrition and quality. However, recent research in geospatial data collection and analysis as well as intelligent deep learning will help us study this further.

Employing the efficiency and interconnection of computer vision and geospatial technology, we want to study whether healthy food in the community is attainable. Specifically, with the help of deep learning in …

Universal Image Segmentation For Optical Identification Of 2d Materials, Joshua Island, Randy M. Sterbentz, Kristine L. Haley

Physics & Astronomy Faculty Research

Machine learning methods are changing the way data is analyzed. One of the most powerful and widespread applications of these techniques is in image segmentation wherein disparate objects of a digital image are partitioned and classified. Here we present an image segmentation program incorporating a series of unsupervised clustering algorithms for the automatic thickness identification of two-dimensional materials from digital optical microscopy images. The program identifies mono- and few-layer flakes of a variety of materials on both opaque and transparent substrates with a pixel accuracy of roughly 95%. Contrasting with previous attempts, application generality is achieved through preservation and analysis …

Infrared Image Segmentation Of Aircraft Skin Based On Otsu And Improved I-Ching Divination Evolutionary Algorithm, Kun Wang, Ji Yao, Peilun Liu, Wang Li

Journal of System Simulation

Abstract: Infrared thermal imaging non-destructive testing is one of the commonly used methods for aircraft skin detection. Aiming at Otsu's large computational complexity and poor real-time performance, an aircraft skin infrared image segmentation method based on Otsu and an improved I-Ching divination evolutionary algorithm (IDEA) is proposed. The roulette selection operator is improved by using roulette selection for the I-Ching map of state size 3n, from which the n individuals with the maximum fitness values are then selected as new populations. The experimental results show that the proposed algorithm is superior to several other improved optimization algorithms both in terms …

Ship Deck Segmentation In Engineering Document Using Generative Adversarial Networks, Mohammad Shahab Uddin, Raphael Pamie-George, Daron Wilkins, Andres Sousa Poza, Mustafa Canan, Samuel Kovacic, Jiang Li

Engineering Management & Systems Engineering Faculty Publications

Generative adversarial networks (GANs) have become very popular in recent years. GANs have proved to be successful in different computer vision tasks including image-translation, image super-resolution etc. In this paper, we have used GAN models for ship deck segmentation. We have used 2D scanned raster images of ship decks provided by US Navy Military Sealift Command (MSC) to extract necessary information including ship walls, objects etc. Our segmentation results will be helpful to get vector and 3D image of a ship that can be later used for maintenance of the ship. We applied the trained models to engineering documents provided …

Impact Of Image Segmentation Techniques On Celiac Disease Classification Usingscale Invariant Texture Descriptors For Standard Flexible Endoscopic Systems, Manarbek Saken, Munkhtsetseg Banzragch Yağci, Nejat Yumuşak

Turkish Journal of Electrical Engineering and Computer Sciences

Celiac disease (CD) is quite common and is a proximal small bowel disease that develops as a permanentintolerance to gluten and other cereal proteins in cereals. It is considered as one of the most di?icult diseases to diagnose.Histopathological evidence of small bowel biopsies taken during endoscopy remains the gold standard for diagnosis.Therefore, computer-aided detection (CAD) systems in endoscopy are a newly emerging technology to enhance thediagnostic accuracy of the disease and to save time and manpower. For this reason, a hybrid machine learning methodshave been applied for the CAD of celiac disease. Firstly, a context-based optimal multilevel thresholding technique wasemployed …

Color-Texture Image Segmentation Approach Designed By Integrating Geodesic Active Contour Model And Multilayer Graph Cut Model, Yang Yong, Liangren Zheng, Guo Ling, Yangdong Ye

Journal of System Simulation

Abstract: An approach of color-texture image segmentation was proposed based on geodesic active contour model and multilayer graph cut model. As the clustering centers were commonly descripted with constant densities, the two phase Chan-vese model was extended to multiphase geodesic active contour model by using the Gaussian distribution to describe the changing density for each phase, and meanwhile the geodesic active contour was added into the proposed model, so that the presented approach could capture out the concave edge. For the minimization of the proposed energy function, a corresponded multilayer graph cut model was designed for resolving global approximated …

Regional Credible Fusion Based Color-Texture Image Segmentation Approach, Yang Yong, Guo Ling, Wenzheng Dai, Yangdong Ye

Journal of System Simulation

Abstract: An approach is proposed based on the combined color information and texture information for color-texture image segmentation. The compacted multi-scale texture information is extracted by using singular value decomposition and principal component analysis dimension reduction under decomposing the multi-scale structure tensor, and then it is integrated with scale information and color information for improving the description ability to color-texture. To avoid the phenomenon such as over-segmentation and error segmentation appeared, the regional credible fusion degree is computed by combining four kinds of region information, such as region adjacency relationship, region size, common edge between regions, and J-divergence distance. Meanwhile, …

Automatic Hair Contour Extraction Method With Complex Background, Yaoli Jin, Shiliang Wang, Xu Gang, Weihua Hu, Yigang Wang

Journal of System Simulation

Abstract: Hair contour extraction has important application in digital entertainment and simulation. An automatic hair contour extraction method for images with complex background was proposed. Image segmentation algorithm from foreground image was used to eliminate the interference of background image; face recognition and hair detection algorithms were employed to find the hair position and realize the automatic identification of the hair; in order to improve the accuracy of contour extraction, adaptive skin detection algorithm was used to eliminate the interference of face. The proposed method was applied on a lot of image examples. Experimental results demonstrate that this method can …

Image Segmentation And Offset Correction Based On Minimal Relative Entropy Theoryand Level Set Method, Xiuqiang Pan, Jinxiao Shan, Caifeng Yang

Journal of System Simulation

Abstract: The new variational level set method is achieved with the combination of the traditional level set method and the energy function which is established by means of statistical model according to the minimal relative entropy.The new method isappliedto object segmentation and offset correction in intensity heterogeneous image.Object segmentation and offset correction are unified according to the evolution of the level set function, anda deviation estimation function with intrinsic smooth feature is obtained.The results prove that the overlapping areas between different tissues are significantly decreased and more accurate results are achieved. In addition, this model is not …

Novel Automatic Pavement Crack Detection Algorithm, Shangbing Gao, Xie Zheng, Zhigeng Pan, Fangzhe Qin, Li Rui

Journal of System Simulation

Abstract: The complexity of noises covers a wide area of actual road images which causes that it is difficult to detect cracks. An automatic pavement crack detection algorithm was proposed in view of the characteristics of crack image in pavement disease. Gray-scale correction and filtering was used to preprocess the crack image. The maximum interclass variance method and Canny operator were used to detect the edge of the disease image, and then the localization and accurate segmentation algorithm was proposed for the crack image based on the maximum connectivity of the crack in the fracture image. The convolution neural network …