Related Experiment Video
Updated: Jun 18, 2026
03:31
End-To-End Deep Neural Network for Salient Object Detection in Complex Environments
Published on: December 15, 2023
Image Deblurring via Frequency-Domain Feature Enhanced Convolutional Neural Networks.
Yecai Guo1,2, Lixiang Ma1,3, Yangyang Zhang2
1School of Electrical and Electronic Engineering, Anhui Institute of Information Technology, Wuhu 241000, China.
Sensors (Basel, Switzerland)
|March 28, 2026
Summary
This study introduces a novel image deblurring algorithm using frequency domain enhancement and convolutional neural networks. The method effectively restores texture details and edges, outperforming others in key metrics.
Area of Science:
- Computer Vision
- Image Processing
- Deep Learning
Background:
- Traditional deblurring methods struggle with texture detail restoration.
- Existing algorithms often exhibit inadequate learning of frequency domain features.
Purpose of the Study:
- To propose an image deblurring algorithm enhancing frequency domain features.
- To improve the restoration of texture details and edges in deblurred images.
Main Methods:
- A Fourier residual module for collaborative spatial and frequency domain feature learning.
- A gated controlled feed-forward unit to boost nonlinear expression.
- An improved supervised attention module in the decoder for feature capture.
- A novel total loss function combining spatial and frequency domain losses.
Main Results:
- Achieved optimal performance in parameter count and running time on GOPRO dataset.
- Obtained high SSIM (0.961) and low LPIPS (0.0278) on GOPRO dataset.
- Achieved high SSIM (0.941) and low LPIPS (0.0286) on HIDE dataset.
- Demonstrated superior detail and edge preservation compared to existing algorithms.
Conclusions:
- The proposed algorithm effectively removes blur while preserving image details and edges.
- The integration of frequency domain enhancement significantly improves deblurring performance.
- The algorithm shows practical value and prospects for computer vision tasks.
Related Concept Videos
01:17Convolution Properties II
652
The important convolution properties include width, area, differentiation, and integration properties.
The width property indicates that if the durations of input signals are T1 and T2, then the width of the output response equals the sum of both durations, irrespective of the shapes of the two functions. For instance, convolving two rectangular pulses with durations of 2 seconds and 1 second results in a function with a width of 3 seconds.
The area property asserts that the area under the...
The width property indicates that if the durations of input signals are T1 and T2, then the width of the output response equals the sum of both durations, irrespective of the shapes of the two functions. For instance, convolving two rectangular pulses with durations of 2 seconds and 1 second results in a function with a width of 3 seconds.
The area property asserts that the area under the...
652
01:20Deconvolution
685
Deconvolution, also known as inverse filtering, is the process of extracting the impulse response from known input and output signals. This technique is vital in scenarios where the system's characteristics are unknown, and they must be inferred from the observable signals.
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
685
01:18Aliasing
776
Accurate signal sampling and reconstruction are crucial in various signal-processing applications. A time-domain signal's spectrum can be revealed using its Fourier transform. When this signal is sampled at a specific frequency, it results in multiple scaled replicas of the original spectrum in the frequency domain. The spacing of these replicas is determined by the sampling frequency.
If the sampling frequency is below the Nyquist rate, these replicas overlap, preventing the original...
If the sampling frequency is below the Nyquist rate, these replicas overlap, preventing the original...
776
01:20Downsampling
777
When considering a sampled sequence with zero values between sampling instants, one can replace it by taking every N-th value of the sequence. At these integer multiples of N, the original and sampled sequences coincide. This process, known as decimation, involves extracting every N-th sample from a sequence, thereby creating a more efficient sequence.
The Fourier transform of the decimated sequence reveals a combination of scaled and shifted versions of the original spectrum. This...
The Fourier transform of the decimated sequence reveals a combination of scaled and shifted versions of the original spectrum. This...
777
01:22Upsampling
696
Managing signal sampling rates is essential in digital signal processing to maintain signal integrity. A decimated signal, characterized by a reduced frequency range due to its lower sampling rate, can be upsampled by inserting zeros between each sample. This upsampling process expands the original spectrum and introduces repeated spectral replicas at intervals dictated by the new Nyquist frequency. To refine this zero-inserted sequence, it is passed through a lowpass filter with a cutoff...
696
01:26Linear Approximation in Frequency Domain
432
Linear systems are characterized by two main properties: superposition and homogeneity. Superposition allows the response to multiple inputs to be the sum of the responses to each individual input. Homogeneity ensures that scaling an input by a scalar results in the response being scaled by the same scalar.
In contrast, nonlinear systems do not inherently possess these properties. However, for small deviations around an operating point, a nonlinear system can often be approximated as linear....
In contrast, nonlinear systems do not inherently possess these properties. However, for small deviations around an operating point, a nonlinear system can often be approximated as linear....
432