Mark Q. Shaw

Automatic Image Segmentation by Dynamic Region Growth and Multiresolution Merging

Image segmentation is a fundamental task in many computer vision applications. In this paper, we propose a new unsupervised color image segmentation algorithm, which exploits the information obtained from detecting edges in color images in the CIE L*a*b* color space. To this effect, by using a color gradient detection technique, pixels without edges are clustered and labeled individually to identify some initial portion of the input image content. Elements that contain higher gradient densities are included by the dynamic generation of clusters as the algorithm progresses. Texture modeling is performed by color quantization and local entropy computation of the quantized image. The obtained texture and color information along with a region growth map consisting of all fully grown regions are used to perform a unique multiresolution merging procedure to blend regions with similar characteristics. Experimental results obtained in comparison to published segmentation techniques demonstrate the performance advantages of the proposed method.

Automatic color image segmentation by dynamic region growth and multimodal merging of color and texture information

Image segmentation is a fundamental task in many computer vision applications. In this paper, we present a novel unsupervised color image segmentation algorithm that utilizes color gradients, dynamic thresholding and texture modeling algorithms in a split and merge framework. To this effect, pixels without edges are clustered and labeled individually to identify the preliminary image content. Pixels that contain higher gradients are further classified by utilizing an iterative dynamic threshold generation technique and an appropriate entropy based texture model. The proposed algorithm was demonstrated successfully on an extensive database of images and benchmarked favorably against prior art.

Unsupervised color image segmentation using a dynamic color gradient thresholding algorithm

We propose a novel algorithm for unsupervised segmentation of color images. The proposed approach utilizes a dynamic color gradient thresholding scheme that guides the region growing process. Given a color image, a weighted vectorbased color gradient map is generated. Seeds are identified and a dynamic threshold is then used to perform reliable growing of regions on the weighted gradient map. Over-segmentation, if any, is addressed by a Similarity Measurebased region merging stage to produce the final segmented image. Comparative results demonstrate the effectiveness of this algorithm for color image segmentation.

Image Classification to Improve Printing Quality of Mixed-Type Documents

Functional image classification is the assignment of different image types to separate classes to optimize their rendering for reading or other specific end task, and is an important area of research in the publishing and multi-Average industries. This paper presents recent research on optimizing the simultaneous classification of documents, photos and logos. Each of these is handled during printing with a class-specific pipeline of image transformation algorithms, and misclassification results in pejorative imaging effects. This paper reports on replacing an existing classifier with a Weka-based classifier that simultaneously improves accuracy (from 85.3% to 90.8%) and performance (from 1458 msec to 418 msec/image). Generic subsampling of the images further improved the performance (to 199 msec/image) with only a modest impact on accuracy (to 90.4%). A staggered subsampling approach, finally, improved both accuracy (to 96.4%) and performance (to 147 msec/image) for the Weka-base classifier. This approach did not appreciable benefit the HP classifier (85.4% accuracy, 497 msec/image). These data indicate staggered subsampling using the optimized Weka classifier substantially improves the classification accuracy and performance without resulting in additional “egregious” misclassifications (assigning photos or logos to the “document” class).

Unsupervised image segmentation by automatic gradient thresholding for dynamic region growth in the CIE L*a*b* color space

In this paper, we propose a novel unsupervised color image segmentation algorithm named GSEG. This Gradient-based SEGmentation method is initialized by a vector gradient calculation in the CIE L*a*b* color space. The obtained gradient map is utilized for initially clustering low gradient content, as well as automatically generating thresholds for a computationally efficient dynamic region growth procedure, to segment regions of subsequent higher gradient densities in the image. The resultant segmentation is combined with an entropy-based texture model in a statistical merging procedure to obtain the final result. Qualitative and quantitative evaluation of our results on several hundred images, utilizing a recently proposed evaluation metric called the Normalized Probabilistic Rand index shows that the GSEG algorithm is robust to various image scenarios and performs favorably against published segmentation techniques.

Plasma etching of polymers like SU8 and BCB

Polymers with high viscosity, like SU8 and BCB, play a dominant role in MEMS application. Their behavior in a well defined etching plasma environment in a RIE mode was investigated. The 40.68 MHz driven bottom electrode generates higher etch rates combined with much lower bias voltages by a factor of ten or a higher efficiency of the plasma with lower damaging of the probe material. The goal was to obtain a well-defined process for the removal and structuring of SU8 and BCB using fluorine/oxygen chemistry, defined using variables like electron density and collision rate. The plasma parameters are measured and varied using a production proven technology called SEERS (Self Excited Electron Resonance Spectroscopy). Depending on application and on Polymer several metals are possible (e.g., gold, aluminum). The characteristic of SU8 and BCB was examined in the case of patterning by dry etching in a CF4/O2 chemistry. Etch profile and etch rate correlate surprisingly well with plasma parameters like electron density and electron collision rate, thus allowing to define to adjust etch structure in situ with the help of plasma parameters.

Assessment of Presence of Isolated Periodic and Aperiodic Bands in Laser Electrophotographic Printer Output

Laser electrophotographic printers are complex systems with many rotating components that are used to advance the media, and facilitate the charging, exposure, development, transfer, fusing, and cleaning steps. Irregularities that are constant along the axial direction of a roller or drum, but which are localized in circumference can give rise to distinct isolated bands in the output print that are constant in the scan direction, and which may or may not be observed to repeat at an interval in the process direction that corresponds to the circumference of the roller or drum that is responsible for the artifact. In this paper, we describe an image processing and analysis pipeline that can effectively assess the presence of isolated periodic and aperiodic bands in the output from laser electrophotographic printers. In our paper, we will discuss in detail the algorithms that comprise the image processing and analysis pipeline, and will illustrate the efficacy of the pipeline with an example.

A general approach for assessment of print quality

Laser electrophotographic printers are complex systems that can generate prints with a number of possible artifacts that are very di_erent in nature. It is a challenging task to develop a single processing algorithm that can effectively identify such a wide range of print quality defects. In this paper, we describe an image processing and analysis pipeline that can effectively assess the presence of a wide range of artifacts, as a general approach. In our paper, we will discuss in detail the algorithm that comprises the image processing and analysis pipeline, and will illustrate the efficacy of the pipeline with a number of examples.

An adaptive and progressive approach for efficient Gradient-based multiresolution color image segmentation

We propose an image segmentation methodology which exploits gradient information in a multiresolution framework. The proposed algorithm commences with a wavelet decomposition procedure to obtain a pyramidal representation of the input image, accompanied by an adaptive threshold generation scheme required for segregating regions of varying gradient densities. At low (coarse) resolution levels, progressive region growth, texture characterization, and region merging modules are integrated together to provide interim segmentations. These interim results are transferred from one resolution level to another as a-priori information, until the final result at the highest (original) resolution is achieved. Performance evaluation on several hundred images demonstrates that our algorithm computationally outperforms various published techniques, with superior segmentation quality.

Local Defect Detection and Print Quality Assessment

Print quality is an important criterion for a printers performance. The detection, classification, and assessment of printing defects can reflect the printers working status and help to locate mechanical problems inside. To handle all these questions, an efficient algorithm is needed to replace the traditionally visual checking method. In this paper, we focus on pages with local defects including gray spots and solid spots. We propose a coarse-to-fine method to detect local defects in a block-wise manner, and aggregate the blockwise attributes to generate the feature vector of the whole test page for a further ranking task. In the detection part, we first select candidate regions by thresholding a single feature. Then more detailed features of candidate blocks are calculated and sent to a decision tree that is previously trained on our training dataset. The final result is given by the decision tree model to control the false alarm rate while maintaining the required miss rate. Our algorithm is proved to be effective in detecting and classifying local defects compared with previous methods.