Improved Extraction of Objects from Urine Microscopy Images with Unsupervised Thresholding and Supervised U-net Techniques

Aziz, Abdul; Pande, Harshit; Cheluvaraju, Bharath; Dastidar, Tathagato Rai

doi:10.1109/cvprw.2018.00299

Cited by 7 publications

(5 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Neural network based models are non linear models which are able to capture non-linear and complex underlying characteristics of images with high degree of accuracy, this gives them capability to handle most of segmentation challenges in microorganism images such as, uneven illumination, poor contrast and non uniform undesired small features on microorganim images [81], [103], [106].…”

Section: B Analysis On Machine Learning Based Methodsmentioning

confidence: 99%

“…However, with augmentation technique enough dataset can be generated from few present dataset and applied to neural network models for better segmentation results, for example in [109] and [29] augmentation is applied to increase the dataset for training. Moreover, the application of transfer learning has boosted the use of neural network models in the segmentation of microorganisms, for example the application of transfer learning can be seen in [30] where VGG-16 model is used, in [29] where ResNet50 model is used and in [106], [109], where U-net model is used for segmentation. Therefore, many of the recent works are focusing on application of neural networks.…”

Section: B Analysis On Machine Learning Based Methodsmentioning

confidence: 99%

“…In [106], different segmentation methods are applied on extraction of objects from low contrast urine microscopic images, applied methods are unsupervised method (Sauvola's method as a key component), U-net method, unsupervised with edge thresholding, U-net with edge thresholding. In the experiments, 329 urine sample images are used.…”

Section: A Supervised Machine Learning (Sml)mentioning

confidence: 99%

See 2 more Smart Citations

A State-of-the-Art Survey for Microorganism Image Segmentation Methods and Future Potential

Kulwa

Zhao

et al. 2019

IEEE Access

View full text Add to dashboard Cite

Microorganisms play a great role in ecosystem, wastewater treatment, monitoring of environmental changes, and decomposition of waste materials. However, some of them are harmful to humans and animals such as tuberculosis bacteria and plasmodium. In such course, it is important to identify, track, analyze, consider the beneficial side and get rid of the negative effects of microorganisms using fast, accurate, and reliable methods. In recent decades, image analysis techniques have been used to address the drawbacks of manual traditional approaches in the identification and analysis of microorganisms. As image segmentation being an important step (technique) in the detection, tracking, monitoring, feature extraction, modeling, and analysis of microorganisms, different methods have been deployed, from classical approaches to current deep neural networks upon different challenges on microorganism images. This survey comprehensively analyses the various studies focused on developing microorganism image segmentation methods in the last 30 years (since 1989). In this survey, segmentation methods are categorized into classical and machine learning methods. Furthermore, these methods are subcategorized into threshold-based, regionbased, and edge-based which belong to classical methods, supervised and unsupervised machine leaningbased methods which belong to machine learning category. A growth trend of different methods and most frequently used methods in each category are meticulously analyzed. A clear explanation of the suitability of these methods for different segmentation challenges encountered on microscopic microorganism images is also enlightened. INDEX TERMSMicroorganism segmentation, content-based microscopic image analysis, feature extraction, microscopic images, classical methods, machine learning.

show abstract

Section: B Analysis On Machine Learning Based Methodsmentioning

confidence: 99%

Section: B Analysis On Machine Learning Based Methodsmentioning

confidence: 99%

Section: A Supervised Machine Learning (Sml)mentioning

confidence: 99%

See 1 more Smart Citation

A State-of-the-Art Survey for Microorganism Image Segmentation Methods and Future Potential

Kulwa

Zhao

et al. 2019

IEEE Access

View full text Add to dashboard Cite

show abstract

“…People begin to use the deep learning methods to extract objects or recognize urine sediment particles. Aziz et al [34] proposed a novel unsupervised method for extracting objects from urine sediment images and applied U-net for extracting these objects. Liang et al [35] proposed feature pyramid network with denseNet (DFPN) method to solve the problem of category confusion in the urine sediment images, which can classify 7 categories.…”

Section: B Automatic Recognition Of Urine Sediment Imagesmentioning

confidence: 99%

Research on Urine Sediment Images Recognition Based on Deep Learning

et al. 2019

IEEE Access

View full text Add to dashboard Cite

Detection of urine sediment microscopic images of human urine samples plays an important part in vitro examination. Doctors usually use automatic urine sediment analyzer to assist manual examine. At present, automatic urine sediment analyzers mostly use traditional method of artificial feature extraction to recognize urine sediment images. However, traditional image processing methods based on the selection and combination of feature operators and classifiers require a lot of work and subjective experience for engineers in the implementation process. It's also difficult to deal with urine sediment images recognition tasks with large scale categories, and particles in some different categories are often confused in recognition using traditional image processing methods, such as red blood cells (RBCs) and white blood cells (WBCs). In this paper, a combination convolution neural network (CNN) recognition method with area feature algorithm is proposed. The disadvantage that CNN can weaken the area feature of input image is solved by area feature algorithm (AFA) proposed in this paper. The network models which use 300,000 urine sediment images for training can quickly and accurately recognize 10 categories of urine sediment images, and several confusing categories' recognition indexes are remarkably improved. The test accuracy in the test set reached 97%. INDEX TERMS Automatic diagnosis, deep learning, image processing, urine sediment.

show abstract

“…This aspect is crucial, as large databases of labeled data are typically not readily available for most scientific imaging applications. U-Net has been applied to various datasets, such as urine microscopic images (Aziz et al, 2018), ADF-STEM images (Ge & Xin, 2018), corneal endothelial cell images (Daniel et al, 2019), and fluorescently labeled cell nuclei images (Gudla et al, 2019). Many other works performed similar microscopy segmentation tasks on the nanoscale using modified versions of the U-Net Architecture such as EM-Net (Khadangi et al, 2020), Fully Residual U-Net (Gómez-de Mariscal et al, 2019), Inception U-Net (Punn & Agarwal, 2020), and the domain adaptive approach with two coupled U-Nets (Bermúdez-Chacón et al, 2018).…”

Section: Introductionmentioning

confidence: 99%

Deep Learning Segmentation of Complex Features in Atomic-Resolution Phase-Contrast Transmission Electron Microscopy Images

et al. 2021

View full text Add to dashboard Cite

show abstract

Improved Extraction of Objects from Urine Microscopy Images with Unsupervised Thresholding and Supervised U-net Techniques

Cited by 7 publications

References 19 publications

A State-of-the-Art Survey for Microorganism Image Segmentation Methods and Future Potential

A State-of-the-Art Survey for Microorganism Image Segmentation Methods and Future Potential

Research on Urine Sediment Images Recognition Based on Deep Learning

Deep Learning Segmentation of Complex Features in Atomic-Resolution Phase-Contrast Transmission Electron Microscopy Images

Contact Info

Product

Resources

About