Effects of JPEG and JPEG2000 Lossy Compression on Remote Sensing Image Classification for Mapping Crops and Forest Areas

Zabala, Alaitz; Pons, Xavier; Dı́az-Delgado, Ricardo; García, Fernando; Aulí-Llinàs, Francesc; Serra-Sagristà, Joan

doi:10.1109/igarss.2006.203

Cited by 24 publications

(26 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is important to stress that the goal is not to transmit images that might be visually pleasing, but rather ones that contain enough information to perform the task (Figure 4). Previous studies [16][17][18] have shown mixed effects of compression on image classification. In this study, classification accuracy was shown to improve with increased image down-sampling and/or compression more universally even than in Ref.…”

Section: Discussionmentioning

confidence: 99%

Authentic Versus Counterfeit Image Classification after Re-Sampling and Compression

Simske¹,

Sturgill²,

Aronoff³

2010

jist

View full text Add to dashboard Cite

Image-based applications such as inspection, quality assurance, print quality assessment, and authentication are often bandwidth limited, and benefit from image down-sampling or compression. Often the down-sampling or compression strategy is selected without considering the impact on the functional goal of the imaging task. In this paper, we use a specific "functional" imaging application-the distinction between images of authentic products and counterfeit products-to assess the impact of down-sampling and compression on the classification accuracy of the counterfeit detection imaging software. Our samples are originally scanned at high resolution (600 dots/in, or dpi) and then down-sampled to as little as 10 ppi (continuous tone pixels/in). JPEG compression to 2% and 1% of the original image size were performed separately or in combination with the various down-sampling factors. A simple Gaussian classifier was used throughout. We found that improved classification accuracy was obtained even for samples compressed or down-sampled by more than 99%. The implications of these findings for improved inspection results, authentication accuracy and counterfeit detection are then discussed.

show abstract

Section: Discussionmentioning

confidence: 99%

Authentic Versus Counterfeit Image Classification after Re-Sampling and Compression

Simske¹,

Sturgill²,

Aronoff³

2010

jist

View full text Add to dashboard Cite

show abstract

“…Their general conclusion is that if the compressed imaged size is 10% of the original image size there is no influence on image quality or on the quality of the DEM generated. Regarding thematic quality, results obtained in classifications in previous studies where not conclusive 7,8,9,10 but recent ones conclude that a compression ratio of 0.4 can be used in the worse case 11 .…”

Section: Introductionmentioning

confidence: 95%

Implications of JPEG2000 lossy compression on multiple regression modelling

et al. 2007

View full text Add to dashboard Cite

Multiple regression is a common technique used when performing digital analysis on images to obtain continuous, quantitative, variables (as temperature, biomass, etc). In this scenario pixels are treated as samples from which several independent variables are known; when remote sensing images are available, the different spectral bands offered by a given sensor are often used as independent variables. The dependent variable is also a quantitative variable, such as a forest inventory variable or a climate variable (e.g., temperature). This paper presents an evaluation of the implications of JPEG2000 lossy compression when applied to these regression processes. Annual minimum and annual mean air temperature are modelled using two methods according to the independent variables used: only geographical, and geographical and remote sensing images as independent variables. Raster matrix representing independent variables were compressed using compression ratios from 50% up to 0.01% of the original file size. Results have revealed that, even at high compression ratios, practically the same accuracy, measured with independent reference climatic stations, is obtained, so JPEG2000 seems an interesting technique not heavily affecting these common modelling approaches.

show abstract

“…Recent studies have demonstrated that, for the majority of applications, a certain degree of lossy compression might be applied without affecting the results of final applications: digital elevation model generation [13,14], and digital image classification [15][16][17].…”

Section: Introductionmentioning

confidence: 99%

JPEG2000 encoding of images with NODATA regions for remote sensing applications

Zabala

González-Conejero

Serra-Sagristà

et al. 2010

J. Appl. Remote Sens

View full text Add to dashboard Cite

The aim of this work is to, within the JPEG2000 framework, enhance the coding performance obtained for images that contain regions without useful information, or without information at all, here named as NODATA regions. In Geographic Information Systems (GIS) and in Remote Sensing (RS), NODATA regions arise due to several factors, such as geometric and radiometric corrections, atmospheric events, the overlapping of successive layers of information, etc. Most coding systems are not devised to consider these regions separately from the rest of the image, sometimes causing a loss in the coding efficiency and in the post-processing applications. We propose two approaches that address this issue; the first technique (Average Data Region, ADR) is carried out as simple pre-processing and the second technique (ShapeAdaptive JPEG2000, SA-JPEG2000) modifies the coding system to avoid the regions without information. Experimental results, performed on data from real applications and different scenarios, suggest that the proposed approaches can achieve, e.g., for SA-JPEG2000, a Signal-toNoise Ratio improvement of about 8 dB. Moreover, in a post-processing application such as a digital classification, the best classification results are obtained when the proposed approaches SA-JPEG2000 and ADR are applied.

show abstract

Effects of JPEG and JPEG2000 Lossy Compression on Remote Sensing Image Classification for Mapping Crops and Forest Areas

Cited by 24 publications

References 4 publications

Authentic Versus Counterfeit Image Classification after Re-Sampling and Compression

Authentic Versus Counterfeit Image Classification after Re-Sampling and Compression

Implications of JPEG2000 lossy compression on multiple regression modelling

JPEG2000 encoding of images with NODATA regions for remote sensing applications

Contact Info

Product

Resources

About