Boundary localization in texture segmentation

Localizing boundaries between textured image regions without sacrificing the labeling accuracy of interior regions remains a problem in segmentation. Difficulties arise because of the conflicting requirements of localization and labeling. Boundary localization usually demands observing the features over small neighborhoods, whereas labeling accuracy increases with the size of the observation neighborhood. This problem is further exacerbated in texture segmentation by the spatially distributed nature of texture features. In this correspondence, we develop a multiresolution approach that combines localized and distributed features to directly address boundary localization in texture segmentation. Maximum localization is achieved by using the gray-level discontinuities at the boundary between textures to define the boundary. The properties that characterize the gray-level discontinuity at texture boundaries are developed and an algorithm is designed to localize the boundary using these discontinuities. This segmentation algorithm is implemented and successfully tested on a set of Brodatz texture mosaics and AVHRR satellite imagery.     

Reduction of noise in AVHRR channel 3 data with minimum distortion

The channel 3 data of the Advanced Very High Resolution Radiometer (AVHRR) on the NOAA series of weather satellites (NOAA 6-12) are contaminated by instrumentation noise. The signal to noise ratio (S/N) varies considerably from image to image and the between sensor variation in S/N can be large. The characteristics of the channel noise in the image data are examined using Fourier techniques. A Wiener filtering technique is developed to reduce the noise in the channel 3 image data. The noise and signal power spectra for the Wiener filter are estimated from the channel 3 and channel 4 AVHRR data in a manner which makes the filter adaptive to observed variations in the noise power spectra. Thus, the degree of filtering is dependent upon the level of noise in the original data and the filter is adaptive to variations in noise characteristics. Use of the filtered data to improve image segmentation, labeling in cloud screening algorithms for AVHRR data, and multichannel sea surface temperature (MCSST) estimates is demonstrated. Examples also show that the method can be used with success in land applications. The Wiener filtering model is compared with alternate filtering methods and is shown to be superior in all applications tested     

Application of neural networks to AVHRR cloud segmentation

The application of neural networks to cloud screening of AVHRR data over the ocean is investigated. Two approaches are considered, interactive cloud screening and automated cloud screening. In interactive cloud screening a neural network is trained on a set of data points which are interactively selected from the image to be screened. Because the data variability is limited within a single image, a very simple neural network topology is sufficient to generate an effective cloud screen. Consequently, network training is very quick and only a few training samples are required. In automated cloud screening, where a general network is designed to handle all images, the data variability can be significant and the resulting neural network topology is more complex. The latitudinal, seasonal and spatial dependence of cloud screening large AVHRR data sets is studied using an extensive data set spanning 7 years. A neural network and associated feature set are designed to cloud screen this data set. The sensitivity of the thermal infrared bands to high atmospheric water vapor concentration was found to limit the accuracy of cloud screening methods which rely solely on data from these channels. These limitations are removed when the visible channel data is used in combination with the thermal infrared data. A post processing algorithm is developed to improve the cloud screening results of the network in the presence of high atmospheric water vapor concentration. Post processing also is effective in identifying pixels contaminated by subpixel clouds and/or amplifier hysteresis effects at cloud-ocean boundaries. The neural network, when combined with the post processing algorithm, produces accurate cloud screens for the large, regionally distributed AVHRR data set