How can Fourier descriptors be used for image retrieval?
Best insight from top research papers
Fourier descriptors can be used for image retrieval by extracting shape-based features from images and using them to calculate similarity. The starting point of the contour is determined using methods such as the minimum inertia axis . The Fourier transform is then applied to obtain eigenvectors, which are used to describe the shape of the image . Similarity between images is measured using techniques such as the correlation coefficient or Euclidean distance . This method has been applied in various domains, including medical imaging for the retrieval of nose disease images . Weighted distances can also be used to compute shape similarity, taking into account the significance of different harmonics in the Fourier descriptors . Overall, Fourier descriptors provide an efficient and effective way to describe and retrieve images based on their shape features.
Answers from top 5 papers
More filters
| Papers (5) | Insight |
|---|---|
| The provided paper does not mention the use of Fourier descriptors for image retrieval. | |
| The paper does not mention how Fourier descriptors are used for image retrieval. | |
30 Jul 2015 | The paper explains that Fourier descriptors can be used for image retrieval by computing the similarity between descriptors using measures such as City-block or Euclidean distance. |
12 Jan 2019 | Fourier descriptors can be used for image retrieval by extracting the edge of the image, determining the starting point of the contour using the minimum inertia axis, and then using Fourier transform to obtain eigenvectors for similarity calculation. |
25 Dec 2016 | Fourier descriptors are used for shape-based feature extraction in the paper. These features are extracted from input images and stored in a database for comparison with query features to retrieve similar images. |
Related Questions
What are some practical applications of the Fourier Transform in signal processing?5 answersThe Fourier Transform finds practical applications in various signal processing domains. In Digital Signal Processing (DSP), it aids in analyzing and enhancing signals like audio music by breaking them down into frequency components. Additionally, the Fourier Transform is crucial in fields like mobile telephony, mobile internet, and automatic control in modern appliances and transport systems, motivating readers to grasp its basics. Moreover, it plays a significant role in minimizing hardware costs for implementing frequency selection algorithms on programmable logic devices without multiplication operations, through the multi-stage discrete Fourier transform methodology. Furthermore, in biomedical signal processing, Fourier Transform is utilized alongside Wavelet Transform to process low-frequency signals like ECG and PPG for accurate clinical monitoring and diagnosis.
What is signifance of fourier series?5 answersThe significance of Fourier series lies in its diverse applications across various fields. In mathematics, Fourier series are essential for representing periodic functions through a combination of sinusoidal and cosinusoidal terms, providing a comprehensive insight into the harmonic structure of signals. Additionally, Fourier series play a crucial role in cryptography, where they are utilized in derivation functions for processes like authentication, verification, and digital signatures, leveraging prime numbers for initial value computation. Moreover, in orthodontics, Fourier analysis enables the accurate quantification of mandibular shape, offering a detailed representation with minimal coefficients and aiding in treatment planning by identifying distinct growth characteristics and functional patterns. Overall, Fourier series serve as a fundamental tool in mathematics, signal processing, cryptography, and medical fields, showcasing their versatility and importance in various applications.
How does the Fourier transform help in image processing?4 answersThe Fourier transform is a fundamental tool in image processing that is used for various purposes such as filtering, analysis, reconstruction, and compression of images. It allows us to transform a time-domain signal into its frequency-domain representation, enabling us to analyze the different frequency components present in an image. This analysis can be used for tasks like image filtering, where specific frequency components can be enhanced or suppressed to achieve desired effects. Additionally, the Fourier transform can be used for image reconstruction, where algorithms supported by the transform are employed to reconstruct images from incomplete or degraded data. The performance of different reconstruction algorithms can be evaluated using image quality assurance metrics like MSE, PSNR, SNR, SSIM, and NIQE. The Fourier transform also finds applications in tasks like image completion and classification, where it allows us to work in Fourier space, which is inaccessible to convolutional architectures.
Double image optical encryption fourier transform?4 answersDouble image optical encryption using Fourier transform is a technique that enhances the security of image encryption systems. Several papers propose different methods for achieving this. Perez et al. present a nonlinear joint transform correlator (JTC) architecture in the Fourier domain (FD) for encrypting and decrypting two simultaneous images. Chen et al. propose a double encryption system based on near-field Fourier ptychography (FP) that utilizes a spatial light modulator to control the displacement of the object in the spatial domain, resulting in higher concealment and security. Faragallah et al. introduce two optical double image encryption systems using the Fresnel Transform (FrT) and chaotic maps, which combine the advantages of noise immunity from chaotic maps and security encryption from FrT. Qiu et al. design a double-image encryption algorithm that combines fractional Fourier transform with discrete fractional angular transform, effectively reducing the correlation among adjacent pixels of the plaintext images. Zhou et al. propose a double-image encryption algorithm based on the phase-truncated multiple-parameter Fresnel transform, which enhances security through pixel scrambling, diffusion, and phase truncation.
What are the limitations of Fourier modulation in tracking and fast imaging of a translational object?5 answersFourier modulation has limitations in tracking and fast imaging of a translational object. One limitation is that conventional background subtraction methods used in Fourier reconstruction schemes suffer from slow updating of environmental changes and cannot accurately extract the boundaries of moving objects. Another limitation is that Fourier imaging systems require multiple grid pairs, which can be expensive to produce, in order to create rudimentary Fourier derived images. Additionally, Fourier modulation methods based on single-pixel detection can achieve high frame rates for tracking, but they cannot simultaneously image the object. These limitations highlight the need for improved methods in motion detection and imaging of moving objects.
How can the Fourier integral be used to solve partial differential equations?5 answersThe Fourier integral can be used to solve partial differential equations by applying the Fourier transform to both sides of the equation. This allows for the equation to be transformed into an algebraic equation in the frequency domain, which can be easier to solve. The Fourier method can be used to obtain classical solutions for initial-boundary value problems for first-order partial differential equations with involution in the function and its derivative. The series produced by the Fourier method can be represented as a formal solution of the problem, with the sum S0 explicitly calculated. Refined asymptotic formulas for the solution of the Dirac system can be used to show that the series obtained from the formal solution converges uniformly. The Fourier integral is also useful in solving partial differential equations by either Rayleigh-Ritz or collocation techniques, using compactly supported radial basis functions.