Table 1 Comparative analysis of image enhancement techniques in medical imaging (2022–2025).

From: Transfer learning with fuzzy decision support for multi-class lung disease classification: performance analysis of pre-trained CNN models

Enhancement method

Key advantage

Limitation

Clinical suitability

References

Histogram equalization (HE)

Improves global contrast

Amplifies noise in uniform areas

Limited

33

CLAHE

Local contrast preservation

Moderate computational load

High

33

Wavelet-based enhancement

Strong edge preservation

High computational cost

Moderate

34

Hybrid multi-technique

Balanced contrast and denoising

Requires careful tuning

Very High

35

Diffusion/transformer-based

Superior low-light performance

Computationally intensive

Excellent (Future)

36

GAN-based enhancement

Learns task-specific features

High training complexity

Excellent

37

Back to article page