Color Space Transformation and Multi-Class Weighted Loss for Adhesive White Blood Cell Segmentation

Li, Huiying; Zhao, Xiaoqing; Su, Anyang; Zhang, Haitao; Liu, Jingxin; Gu, Gui‑ying

doi:10.1109/access.2020.2970485

Cited by 29 publications

(19 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Their results showed that the highest segmentation accuracy was 96.92% using the S component of the HSV color space. Moreover, Li et al [56] proposed a weighted cross-entropy loss function based on class weight and distance transformation weight for deep learning using U-Net for the WBC segmentation using ALL-IDB1, which resulted in 94.92% segmentation accuracy without data augmentation. In addition, our counting accuracy was 93.7% using the entire ALL-IDB dataset, which is 12.7% higher than the accuracy reported by Mahmood et al [57], where color space conversion and Hough transform were used for WBC detection.…”

Section: Discussionmentioning

confidence: 99%

A Cascaded Classification-Segmentation Reversible System for Computer-Aided Detection and Cells Counting in Microscopic Peripheral Blood Smear Basophils and Eosinophils Images

2021

View full text Add to dashboard Cite

Computer-aided image analysis has a pivotal role in automated counting and classification of white blood cells (WBCs) in peripheral blood images. Due to their different characteristics, our proposed approach is based on investigating the variations between the basophils and eosinophils in terms of their color histogram, size, and shape before performing the segmentation process. Accordingly, we proposed a cascaded system using a classification-based segmentation process, called classification-segmentation reversible system (CSRS). Prior to applying the CSRS system, a Histogram-based Object to Background Disparity (HOBD) metric was deduced to determine the most appropriate color plane for performing the initial WBC detection (first segmentation). Investigating the local histogram features of both classes resulted in a 92.4% initial classification accuracy using the third-degree polynomial support vector machine (SVM) method. Subsequently, in the proposed CSRS approach, transformation-based segmentation algorithms were developed to fit the specific requirements of each of the two predicted classes. The proposed CSRS system is used, where the images from an initial classification process are fed into a second segmentation process for each class separately. The segmentation results demonstrated a similarity index of 94.9% for basophils, and 94.1% for eosinophils. Moreover, an average counting accuracy of 97.4% for both classes was achieved. In addition, a second classification was carried out after applying the CSRS, achieving a 5.2% increase in accuracy compared to the initial classification process.

show abstract

Section: Discussionmentioning

confidence: 99%

A Cascaded Classification-Segmentation Reversible System for Computer-Aided Detection and Cells Counting in Microscopic Peripheral Blood Smear Basophils and Eosinophils Images

2021

View full text Add to dashboard Cite

show abstract

“…Hence, the developed GFS Extreme-Net model consumes more time for labeling, and also it is very difficult to identify the specific extreme points that reflects the best geometric features of a target. Additionally, H. Li, X. Zhao, A. Su, H. Zhang, J. Liu, and G. Gu, [17] developed a weight map on the basis of distance transformation weight and class weight to improve the ability of loss function in U-Net for effectively learning the cell border feature. The experimental results showed that the developed model achieved better performance in white blood cell segmentation on the ALL-IDB1 database.…”

Section: Literature Surveymentioning

confidence: 99%

Edge Boost Curve Transform and Modified ReliefF Algorithm for Communicable and Non Communicable Disease Detection Using Pathology Images

Reddy¹,

Channegowda²

2021

IJIES

View full text Add to dashboard Cite

In this paper, a five phase model is proposed for early detection of communicable and non-communicable diseases like Haemoprotozoan and breast cancer using pathology images. At first, color normalization technique is utilized to improve the visual quality of the collected histology images. Next, edge boost curve transform is employed to segment nuclei and non-nuclei cells from the enhanced images. The developed segmentation methodology delivers good results in overlapped database. Further, the segmented image is converted into one dimensional vectors and then modified reliefF algorithm is applied to choose the active feature vectors to achieve better classification. Finally, deep neural network is accomplished to classify the Haemoprotozoan images as anaplasmosis, babesiosis and theileriosis, and breast images as malignant or benign. From the experimental result, the proposed model; modified reliefF-deep neural network obtained maximum classification accuracy of 97.6% in Haemoprotozoan disease detection and 95.94% in breast cancer detection, which are better related to other comparative techniques like Random Forest, Multi Support Vector Machine and K-Nearest Neighbor.

show abstract

“…Therefore, the HSV color information [22] is used to process wood images before classification. The wood color ranges in the H, S, and V channels can be used for removing the background and stains simultaneously [23].…”

Section: Image Preprocessingmentioning

confidence: 99%

Color Classification of Wooden Boards Based on Machine Vision and the Clustering Algorithm

et al. 2020

View full text Add to dashboard Cite

Color classification of wooden boards is helpful to improve the appearance of wooden furniture that is spliced from multiple wooden boards. Due to the similarity of colors among wooden boards, manual color classification is inaccurate and unstable. Thus, supervised learning algorithms can hardly be used in this scenario. Moreover, wooden boards are long, and their images have a high resolution, which may lead to the growth of computational complexity. To overcome these challenges, in this paper, we propose a new mechanism for color classification of wooden boards based on machine vision. The image of the wooden board is preprocessed to subtract irrelevant colors, and the feature vector is extracted based on 3D color histogram to reduce the computational complexity. In the offline clustering, the feature vector sets are partitioned into different clusters through the K-means algorithm. Then, the clustering result can be used in the online classification to classify the new wood image. Furthermore, to process the abnormal images of wooden boards, we propose an improved algorithm with centroid improvement and image filtering. The experimental results verify the effectiveness of the proposed mechanism.

show abstract

Color Space Transformation and Multi-Class Weighted Loss for Adhesive White Blood Cell Segmentation

Cited by 29 publications

References 43 publications

A Cascaded Classification-Segmentation Reversible System for Computer-Aided Detection and Cells Counting in Microscopic Peripheral Blood Smear Basophils and Eosinophils Images

A Cascaded Classification-Segmentation Reversible System for Computer-Aided Detection and Cells Counting in Microscopic Peripheral Blood Smear Basophils and Eosinophils Images

Edge Boost Curve Transform and Modified ReliefF Algorithm for Communicable and Non Communicable Disease Detection Using Pathology Images

Color Classification of Wooden Boards Based on Machine Vision and the Clustering Algorithm

Contact Info

Product

Resources

About