Image recognition in single-scale and multiscale decoders

Schilling, D.; Cosman, Pamela C.; Berry, Craig

doi:10.1109/acssc.1998.750909

Cited by 6 publications

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Methods (11) (12) showed that human observers can recognize and respond to image contents at bit-rates of 0.05 to 0.1 bpp, using the SPIHT algorithm, for many natural image classes. These results suggest that it is important for algorithm designers to consider very low bit rates for fast browsing tasks and wireless Internet access.…”

Section: Resultsmentioning

confidence: 99%

Post-Processing for Restoring Edges and Removing Artifacts of Low Bit Rates Wavelet-Based Image

Lü

Sekiya

et al. 2007

IEEJ Trans.EIS

View full text Add to dashboard Cite

Section: Resultsmentioning

confidence: 99%

Post-Processing for Restoring Edges and Removing Artifacts of Low Bit Rates Wavelet-Based Image

Lü

Sekiya

et al. 2007

IEEJ Trans.EIS

View full text Add to dashboard Cite

“…In this work we are primarily interested in the low end of the bit rate progression, where recognition is likely to take place, rather than the high-rate, high-quality end. Our work in [20], [21] showed that human observers can recognize and respond to image content at bit rates of 0.05 to 0.1 bpp, using the SPIHT algorithm, for many natural image classes. These results suggest that it is important for algorithm designers to consider very low bit rates when designing compression algorithms for fast browsing tasks and wireless Internet access.…”

Section: Resultsmentioning

confidence: 99%

Preserving step edges in low bit rate progressive image compression

Schilling

Cosman²

2003

IEEE Trans. on Image Process.

Self Cite

View full text Add to dashboard Cite

With the growing importance of low-bandwidth applications, such as wireless access to the Internet, images are often sent or received at low bit rates. At these bit rates, they suffer from significant distortion and artifacts, making it difficult for those viewing the images to understand them. We present two progressive compression algorithms that focus on preserving the clarity of important image features, such as edges, at compression ratios of 80:1 and more. Both algorithms capture and encode the locations of important edges in the images. The first algorithm then transmits a standard SPIHT (set partitioning in hierarchical trees) bit stream, and at the decoder applies a nonlinear edge-enhancement procedure to improve the clarity of the encoded edges. The second approach uses a modified wavelet transform to "remove" the edges, and encodes the remaining texture information using SPIHT. With both approaches, features in the images that may be important for recognition are well preserved, even at low bit rates.

show abstract

“…The approach established by this investigation promises improvement upon state of the art wavelet decompression techniques for a wide range of applications, including audio denoising (Schremmer, Haenselmann, and Bömers 2001), sea clutter noise reduction for radar proximity fusing (Noel and Szu 1998), signal compression (Saito 1994), object detection (Zhu and Schwartz 2002), fingerprint compression (Bradley, Brislawn, and Hopper 1993), image denoising (Chang, Yu, and Vetterli 1998) , image enhancement (Wang, Wu, Castleman, and Xiong 2001), image recognition (Schilling, Cosman, and Berry 1998), and speech recognition (Long and Datta 1998).…”

Section: Applications Of the New Technology And Future Researchmentioning

confidence: 99%

Adaptive Filtering in the Wavelet Transform Domain Via Genetic Algorithms

Moore¹,

Marshall²,

Balster³

2004

View full text Add to dashboard Cite

The public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Department of Defense, Washington Headquarters Services, Directorate for Information SPONSORING/MONITORING AGENCY REPORT NUMBER(S) AFRL-IF-WP-TR-2004-1553 DISTRIBUTION/AVAILABILITY STATEMENTApproved for public release; distribution is unlimited. SUPPLEMENTARY NOTESReport contains color. ABSTRACTThis report primarily addresses the problem of quantization noise since it is one of the very few processes that potentially eliminates valuable signal information. In a lossy compression system, the quantization step is totally responsible for information loss resulting in quality reduction of the reconstructed signal. For this effort, adaptive filtering techniques are utilized for modifying standard, "off-the-shelf," discrete wavelet transform (DWT) coefficients in the hopes that the differential errors between the original uncorrupted signal and the corrupted signal will be minimized. These results show that coefficients evolved by a genetic algorithm (GA) can indeed outperform standard wavelet coefficients for reconstruction of one-and two-dimensional data subjected to quantization. This approach consistently identified coefficient sets that reduced mean squared error (MSE) and improved peak signal-to-noise ratio (PSNR) for inverse DWTs. SUBJECT TERMSGenetic algorithms, discrete wavelet algorithms, quantization noise.

show abstract

Image recognition in single-scale and multiscale decoders

Abstract: Abstract

Cited by 6 publications

References 9 publications

Post-Processing for Restoring Edges and Removing Artifacts of Low Bit Rates Wavelet-Based Image

Post-Processing for Restoring Edges and Removing Artifacts of Low Bit Rates Wavelet-Based Image

Preserving step edges in low bit rate progressive image compression

Adaptive Filtering in the Wavelet Transform Domain Via Genetic Algorithms

Contact Info

Product

Resources

About