Recognition of Images Degraded by Gaussian Blur

Flusser, Jan; Farokhi, Sajad; Höschl, Cyril; Suk, Tomáš; Zitová, Barbara; Pedone, Matteo

doi:10.1109/tip.2015.2512108

Cited by 99 publications

(62 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Standard data augmentation techniques such as vertical and horizontal mirroring, blurring, deformation, and rotation were applied to 50% of the images in each class of the test sets. The deformation was achieved using Moving Least Squares" affine transformation [37] and the blurring by the use of Gaussian function in what is known as Gaussian Blur [38] with a kernel size of 15x15. These steps were necessary to test the ability of the models to generalize on unseen data and also under bad illumination conditions.…”

Section: A Image Acquisition and Preprocessingmentioning

confidence: 99%

Gabor Capsule Network for Plant Disease Detection

Mensah¹,

Weyori²,

Ayidzoe³

2020

IJACSA

View full text Add to dashboard Cite

Crop diseases contribute significantly to food insecurity, malnutrition, and poverty in Africa where the majority of the population is into Agriculture. Manual plant disease recognition methods are widespread but limited, ineffective, costly, and time-consuming making the need to search for automatic and efficient methods of recognition more crucial. Machine learning and Convolutional Neural Networks have been applied in other jurisdictions in an attempt to solve these problems. They have achieved impressive results in this domain but tend to be 'data-hungry', invariant, and vulnerable to attacks that can easily lead to misclassifications. Capsule Networks, on the other hand, avoids the weaknesses of CNNs and has not been widely used in this area. This article, therefore, proposes the use of Gabor and Capsule network to recognize blurred, deformed, and unseen tomato and citrus disease images. Experimental results show that the proposed model can achieve a 98.13% test accuracy which is comparable to the performance of state-of-theart CNN models in the literature. Also, the proposed model outperformed two state-of-the-art deep learning models (which were implemented as baseline models) in terms of robustness, flexibility, fast converges, and having fewer parameters. This work can be extended to other crops and may well serve as a useful tool for the recognition of unseen plant diseases under bad weather and bad illumination conditions.

show abstract

Section: A Image Acquisition and Preprocessingmentioning

confidence: 99%

Gabor Capsule Network for Plant Disease Detection

Mensah¹,

Weyori²,

Ayidzoe³

2020

IJACSA

View full text Add to dashboard Cite

show abstract

“…Citra berupa hasil filter ND dan UV akan diolah terlebih dahulu menjadi sebuah citra grayscale agar citra menjadi bentuk 8 bit. Citra hasil grayscale akan dilakukan Gaussian blur untuk menghilangkan noise sehingga citra seolah lebih halus [12]. Untuk memisahkan antara objek (matahari) dengan background (langit) maka dilakukan proses threshold binary [13].…”

Section: Perancangan Perangkat Lunakunclassified

Pelacak Cahaya Matahari Berbasis Citra pada Panel Surya menggunakan Single Board Computer Lattepanda

Krisandika¹,

Rivai²,

Tasripan³

2019

JTITS

View full text Add to dashboard Cite

Semakin berkurangnya sumber energi yang tidak dapat diperbaharui, maka perlu memanfaatkan sumber energi alternatif yang salah satunya adalah energi matahari. Energi matahari dapat dimaanfatkan dengan menggunakan panel surya. Panel surya hanya mampu mengkonversi sekitar 15 persen dari total energi yang diterima untuk diubah menjadi energi listrik. Salah satu metode untuk meningkatkan efisiensi adalah dengan pelacakan posisi matahari. Pada penelitian ini telah dirancang dan dibuat sistem pelacakan posisi matahari yang berbasis pengolahan citra yang diimplementasikan pada panel surya 50 Watt. Proses ini meliputi konversi citra matahari kedalam bentuk grayscale dan Gaussian blur lalu dilakukan threshold, morphology, dan deteksi kontur untuk mengeliminasi noise. Deteksi lingkaran digunakan untuk mendeteksi posisi matahari pada layar. Proses centeroid dilakukan untuk mendapatkan posisi titik tengah dari matahari yang digunakan untuk mengontrol gerakan 2-axis pada motor DC. Agar gerakan motor lebih presisi maka dilakukan penambahan kontrol proporsional agar panel surya menghadap kearah matahari dengan tepat. Seluruh proses ini dilakukan dengan menggunakan Single Board Computer Lattepanda. Hasil eksperimen menunjukkan bahwa panel surya yang dilengkapi dengan sistem pelacak ini menghasilkan peningkatan daya listrik yang diterima sebesar 6.97 % terhadap panel surya yang diletakkan secara statis.

show abstract

“…3) in the subsystem of image analysis is the selection of local features, such as the end and split for further recognition [9,10] of the fingerprint; 4) in the image analysis subsystem, the obtained parameters of local features are sorted; 5) the recognition subsystem converts the absolute parameters of special points to relative parameters to prevent the effect of parallel transfer and rotation of the finger when scanning the fingerprint [11]; 6) fingerprint recognition with existing fingerprints is based on the relative parameters of each point for each fingerprint stored in the database.…”

Section: Characteristics Of the Functional Structure Of The Systemmentioning

confidence: 99%

Development of the Identification System by Fingerprints

Alanssari¹,

Wahhab²

2018

CTAC&R

View full text Add to dashboard Cite

elements from them, such as a line, area, compact element (letter), and so on. Nowadays, reliable data protection is one of the main criteria by which systems intended for storage and processing of important information should be selected. This is due to the existing risk of unauthorized access to such systems, since they have wide information interaction with related control systems via the INTRERNET. Therefore, information security should be an important part of their development.Security based on the biometric parameters of the human body, in particular fingerprint, has a number of undeniable merits: ease of use, convenience and reliability. The whole process of identification takes little time and requires no effort from those who use the access system.The aim of this work is to develop and implement such a transformation of the image, in which the data on the location of the unique features are stored most fully and with the lowest content of false information [1].The article is structured as follows. Section 1 states the purpose of the study the article justifies expediency of development of the system of identification by fingerprints. Section 2 provides an analytical overview. The method of identification by fingerprints, which has been known for a long time and with the advent of electronic computer technology began to appear software products for the analysis and comparison of images, is described [2].Section 3 sets out the basic system requirements. The creation of an identity recognition system will provide a new opportunity in the field of protection and organization of access to information, as well as the development of new, efficient algorithms for processing raster and converting them to a structural form, to develop a tool that improves the quality of graphical information by reducing distortion and noise. Section 4 describes the main technical solutions of the system project.Dactyloscopy (fingerprint recognition) is the most developed to the date biometric method of personal identification. The catalyst for the development of the method was its widespread use in criminology of the XX century. As each person has a unique papillary pattern of fingerprints, so identification is possible. Typically, algorithms use characteristic points on fingerprints: the end of the pattern line, branching lines, single points. In addition, information about the morphological structure of the fingerprint is attracted: the relative position of the closed lines of the papillary pattern, "arched" and spiral lines. Peculiarities of papillary patterns are converted to some unique codes, which preserves the information content of the fingerprint image. And it is "fingerprint codes" that are stored in the database used for searching and comparing. Currently, fingerprint recognition systems occupy more than half of the biometric market. A lot of companies are engaged in the production of access control systems based on the method of fingerprinting identification. Due to the fact that this direction is one of the oldest, i...

show abstract

Recognition of Images Degraded by Gaussian Blur

Cited by 99 publications

References 59 publications

Gabor Capsule Network for Plant Disease Detection

Gabor Capsule Network for Plant Disease Detection

Pelacak Cahaya Matahari Berbasis Citra pada Panel Surya menggunakan Single Board Computer Lattepanda

Development of the Identification System by Fingerprints

Contact Info

Product

Resources

About