Digital image analysis based automated kiwifruit counting technique

Wijethunga, P.; Samarasinghe, Sandhya; Kulasiri, Don; Woodhead, Ian

doi:10.1109/ivcnz.2008.4762149

Cited by 25 publications

(19 citation statements)

References 8 publications

(8 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A digital camera is used to capture changes in interference colors, and RGB profiles which are evaluated by image analysis techniques [2]. Automated kiwifruit counting system [5] is developed by employing digital image analysis techniques, three simple counting methods and a minimum distance classifier based segmentation technique. An automatic identification and classification tool for the different types of cocci bacterial cells in digital microscopic cell images developed by using image analysis techniques along with K-NN and Neural Network classifiers [6].…”

Section: Advanced Developments In Image Analysismentioning

confidence: 99%

Semantic Image Analysis for Intelligent Image Retrieval

Khodaskar

Ladhake

2015

Procedia Computer Science

View full text Add to dashboard Cite

Image understanding and analysis is the most exciting and fastest-growing research areas in the computer vision. Recent computer vision technologies and algorithms are support efficient semantic image analysis and retrieval. Image analysis is deal with image representation, estimation formula, and sampling density. Image analysis at semantic level is result in automatic extraction of image descriptions as per human perception which ultimately bridge semantic gap between low-level visual features and the high-level concepts capturing the conveyed meaning. Vital semantic image information is basically retrieved from image content, mainly from meaningful image objects and their mutual relations. In this paper, we present Semantic analysis of image by knowledge driven approach , start with Image content analysis with respect to semantic concepts, design image database and knowledge base on the basis of semantic content and retrieval, presentation and modification of image reference database or knowledge base for knowledge delivery intention. Experimental result shows improvement in image retrieval performance and accuracy.

show abstract

Section: Advanced Developments In Image Analysismentioning

confidence: 99%

Semantic Image Analysis for Intelligent Image Retrieval

Khodaskar

Ladhake

2015

Procedia Computer Science

View full text Add to dashboard Cite

show abstract

“…Each set contains five representative images. Images were captured using a custom built acquisition setup used in [5]. The details of imaging conditions for each image set are shown in Table 1.…”

Section: A Data and Pre-processingmentioning

confidence: 99%

“…Images were pre-processed using the steps used in [5]. First, each colour band values were contrast stretched using a nonlinear function, 5 . 0 kx y = , where k is a scalar and x is the input intensity value.…”

Section: A Data and Pre-processingmentioning

confidence: 99%

“…Secondly, the green colour band was omitted and the images were made false colour, replacing R, G and B with B, R and B respectively to 24th International Conference Image and Vision Computing New Zealand (IVCNZ 2009) -63 -achieve better visual contrast. This false colour combination [5] creates an approximately linear relationship between a and b colour components (Fig. 2) in CIELab space, providing a convenient gamut shape to map between images.…”

Section: A Data and Pre-processingmentioning

confidence: 99%

“…Therefore, a generalized model which can detect fruit despite slight variations of the spectral reflectance properties and imaging conditions is of interest to the industry. In this paper, we investigate the potential of developing such a model for detecting kiwifruit within orchards, as an attempt to extend the vision system developed in [5] to be more generalized.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Towards a generalized colour image segmentation for kiwifruit detection

Wijethunga

Samarasinghe

Kulasiri

et al. 2009

2009 24th International Conference Image and Vision Computing New Zealand

Self Cite

View full text Add to dashboard Cite

Abstract-Developing robust computer vision algorithms to detect fruit in trees is challenging due to less controllable conditions, including variation in illumination within an image as well as between image sets. There are two classes of techniques: local-feature-based techniques and shape-based techniques, which have been used extensively in this application domain. Out of the two classes, the local-feature-based techniques have shown higher accuracies over shape-based techniques, but are less desirable due to the requirement of repeated calibration. In this paper, we investigate the potential of developing a generalized colour pixel classifier that can be employed to detect kiwifruit on vines, under variable fruit maturity levels and imaging conditions. First, we observed the colour data patterns of fruit and nonfruit regions from different image sets. With consistant data patterns it was found that a suitable normalization could produce an invariant colour descriptor. Then, a neural network Self-Organizing Map (SOM) model, which has a hierarchical clustering ability was used to investigate the potential of developing a generalized neural network model to classify pixels under variable conditions. Models were built for colour features extracted in CIELab space for both absolute colour values and relative colour descriptors. The paper presents the positive results of the preliminary investigations. The conditions for a successful application of the approach as well as the potential for extending it for automatic calibration will also be discussed.

show abstract

A novel identification method for CPPU‐treated kiwifruits based on images

Yan

2019

J Sci Food Agric

View full text Add to dashboard Cite

BACKGROUND The application of exogenous plant growth regulator, for example forchlorfenuron (CPPU), on kiwifruits has become an important factor that influences kiwifruit economic efficiency and the health development of the kiwifruit industry. Owing to the slight difference in calyx shape between the kiwifruits treated with CPPU (CPPU‐treated kiwifruits) and the kiwifruits without CPPU treatment (CPPU‐untreated kiwifruits), this study aims to provide a cheap, quick, convenient, and non‐destructive method for identifying CPPU‐treated kiwifruits based on the images of kiwifruits captured at visible lights. RESULTS The identification method includes three steps. Firstly, the kiwifruit was extracted from the background by using Otsu algorithm, hole filling operation and ‘bwareaopen’ function. Secondly, the calyx was extracted by using corrosion, image enhancement, hole filling and closing operations. Finally, the length/width ratio of the minimum enclosing rectangle of calyx region was calculated. The kiwifruit was regarded as a CPPU‐treated kiwifruit if the length/width ratio of the rectangle was higher than 1.6. Otherwise, the kiwifruit was regarded as a CPPU‐untreated one. The method had the total identification accuracy rate of 90.0% when the kiwifruit images were captured either by utilizing a smartphone at normal lighting condition or by using an image acquisition system. CONCLUSION The programs run on computer and smartphone were developed, and they could realize kiwifruit identification in 0.6 s and 2 s, respectively. The study makes identifying CPPU‐treated kiwifruits in online processing be realizable, and offers a convenient method for kiwifruit consumers. © 2019 Society of Chemical Industry

show abstract

Digital image analysis based automated kiwifruit counting technique

Abstract: Abstract

Cited by 25 publications

References 8 publications

Semantic Image Analysis for Intelligent Image Retrieval

Semantic Image Analysis for Intelligent Image Retrieval

Towards a generalized colour image segmentation for kiwifruit detection

A novel identification method for CPPU‐treated kiwifruits based on images

Contact Info

Product

Resources

About