Coffee disease detection using a robust HSV color‐based segmentation and transfer learning for use on smartphones

Waldamichael, Fraol Gelana; Debelee, Taye Girma; Ayano, Yehualashet Megersa

doi:10.1002/int.22747

Cited by 25 publications

(10 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Three-dimensional graphics of real surface colors, in brightness, chroma and hue systems (Best, 2017) The effort to extract and process chromaticity data, without any undesired intensity effect of an RGB space, is done for fruit recognition approach using color chromaticity to characterize fruit color with HSV (Garcia et al, 2016). A tradition HSV color space is also used to detect coffee plant diseases, by proposing an HSV color segmentation algorithm to separate the leaves from the background and separate the infected spots on the leaves by automatically identifying the best threshold value for the saturation channel (Waldamichael et al, 2022). Another utilization is to classify images of farmland in different environment, realized using image analysis and classification technology based on the HSV, HSL and HSI color space models (Miao et al, 2015).…”

Section: Related Workmentioning

confidence: 99%

Modification Color Filtering in HSV Color Space

Kusnandar

Santoso

Surendro

2023

Preprint

View full text Add to dashboard Cite

The health condition of plants can be determined by calculating the vegetation index. A healthy plant is identified from photosynthetic activity that reflects a lot of green and absorbs red and blue. Image information from a visible light camera will provide a Red-Green-Blue (RGB) value for each pixel which will be used to calculate the vegetation index. RGB values are largely determined by lighting conditions (scene) at the time of shooting by the camera. Plant images are influenced by high variability of illumination conditions due to changes in natural lighting conditions; bright, cloudy or nearby object illumination. RGB represents a non-uniform color space, thereby needing transformation to other color spaces. The purpose of color space transformation is to replicate the original information from the real world without changing any actual information. Color space transformation often causes information loss, including when mapping colors. The main contribution of this study is the simplification and improvement to the selection speed of desired colors in the HSV color space using the Heaviside step function. Modification was performed by sorting out the green color at a hue value of 90°-150°. The modified results showed that green can be distinguished from other colors. The experimental results on images taken using an unmanned aerial vehicle (UAV) showed that the results of the proposed modification increase the speed of computation time by an average of 16.60%.

show abstract

Section: Related Workmentioning

confidence: 99%

Modification Color Filtering in HSV Color Space

Kusnandar

Santoso

Surendro

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…We took this brand as a reference because it has a complete color range, as many as 69 colors in the form of numbers which are categorized as follows (Figure 5-Figure 9). 20,21,22,23,24,25,28,29,30,31,35,38,43, and 45 (Figure 7).…”

Section: System Designmentioning

confidence: 99%

“…By using HSV Color Extraction, a hue, Saturation, and value-based color selection method. HSV has been used in several previous studies with various purposes, for example, finger detection [26], fruit ripeness [27], coffee disease [28], skin disease [29], land fire detection [30], Irish detection [31], parking detection [32], et cetera. In this research, the application includes foundation and lipstick colors.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

AR Make-up Filter for Social Media using the HSV Color Extraction

Harika

Rachmat

Aulia

et al. 2022

join

View full text Add to dashboard Cite

Choosing the appropriate cosmetics is an arduous task. Because cosmetics are tested directly on the skin to ensure each personâ€™s preferences are met. The consumer repeatedly tries a sample and then discards it until he discovers one that meets his tastes. The cosmetics business and consumers are affected by this move. Companies can utilize Augmented Reality (AR) technology as an alternative to mass-producing cosmetic samples. The difficulty of deploying augmented reality is the difficulty of putting cosmetics into camera video streams. Each individual bears the burden of skin color and its effect on light. HSV Color Extraction was the method employed for this study. The application of augmented reality intends to enable consumers to test cosmetics with their chosen color and assist businesses in competing in the industry by promoting items and engaging customers. This work makes it easier to choose cosmetics using augmented reality and social media. AR simulates the usage of the desired color cosmetics, whereas social media allows users to obtain feedback on their color preferences. The outcomes of this study indicate that augmented reality (AR) apps can display filters in bright, dim, and even wholly dark lighting conditions. This research contributes originality that cosmetic firms can utilize to market their products on social media.

show abstract

“…Machine learning and deep learning methods have found extensive applications in various domains, including disease detection and classification [10][11][12][13][14][15][16][17]. In the context of breast cancer detection and classification, classical machine learning methods have been commonly employed [18,19].…”

Section: Introductionmentioning

confidence: 99%

A Multi-view Breast-wise Abnormality Detection for Breast Cancer Screening using Mammography

Kebede,

Waldamichael,

Debelee

et al. 2023

Preprint

View full text Add to dashboard Cite

Breast cancer has the highest incidence rate among women in Ethiopia compared to other types of cancer. Unfortunately, many cases are detected at a stage where a cure is delayed or not possible. To address this issue, mammography-based screening is widely accepted as an effective technique for early detection. However, the interpretation of mammography images requires experienced radiologists in breast imaging, a resource that is limited in Ethiopia. In this research, we have developed a model to assist radiologists in mass screening for breast abnormalities and prioritizing patients. Our approach combines an ensemble of efficient-based classifiers with YOLOv5, a suspicious mass detection method, to identify abnormalities. The inclusion of YOLOv5 detection is crucial in providing explanations for classifier predictions and improving sensitivity, particularly when the classifier fails to detect abnormalities. To further enhance the screening process, we have also incorporated an abnormality detection model. The classifier model achieves an F1-score of 0.87 and a sensitivity of 0.82. With the addition of suspicious mass detection, sensitivity increases to 0.89, albeit at the expense of a slightly lower F1-score of 0.79.

show abstract

Coffee disease detection using a robust HSV color‐based segmentation and transfer learning for use on smartphones

Cited by 25 publications

References 37 publications

Modification Color Filtering in HSV Color Space

Modification Color Filtering in HSV Color Space

AR Make-up Filter for Social Media using the HSV Color Extraction

A Multi-view Breast-wise Abnormality Detection for Breast Cancer Screening using Mammography

Contact Info

Product

Resources

About