Veterinary Thermographic Image Analysis



[pic]

Veterinary Thermographic Image Analysis

Data and Temperature Normalization

Project Number 7- 64878

Report Number 4878-5

Scott E Umbaugh, BSE, MSEE, PhD

Patrick Solt, BSEE, MSCS, PhD Candidate

Computer Vision and Image Processing Laboratory

Southern Illinois University Edwardsville

August 20, 2008

Summer Semester

Submitted to:

Dr. Catherine Loughin

Dr. Dominic Marino, Chief of Staff

Long Island Veterinary Specialists

TABLE OF CONTENTS

Executive Summary 3

Introduction and Overview 4

Materials and Methods 4

Images 4

Software 4

Data Normalization 7

Color Normalization 8

Experimental Results and Discussion 10

Results from Experiment Set #1. 10

Results from Experiment Set #2. 12

Conclusions 14

Future Work 15

References 16

Appendix – Excel Result File Names 17

Executive Summary

Research and development was undertaken to investigate the application of image analysis and pattern classification to thermographic images of 34 canines. Specifically, thermographic images of canines of the breed Cavalier King Charles Spaniel were examined to investigate the Chiari malformation, or COMS, pathology. Pattern classification algorithms were developed for: severe, moderate and mild classes of the pathology. For these experiments the classification was performed using pairs of images, instead of single images as in previous experiments. The head images were used, and for this report the pairs were: 1) top and left side, 2) top and right side, 3) front and left side, 4) front and right side. The eventual goal for the research is to be able to differentiate normal and abnormal thermographic patterns in canines as a diagnostic and research tool.

Two sets of experiments were performed using two data normalization methods and the original and four color normalization methods for a total for of 20,971,440 experiments. Each experiment consisted of two views of the head, specific classes, a specific data normalization method, a specific color normalization method and specific features. Experiments were performed with all permutations of the nine selected features on each pair of selected body parts. These nine are the five texture features and four of the histogram features – histogram mean was not used. Based on previous experimentation the MinMax and SoftMax data normalization methods provided the best results and were selected for these experiments. The texture distance of 6 provided the best results in previous experimentation and was selected for use here. Regarding color normalization, the original and the four peviously defined methods were used.

The first set of experiments used the clinical images that were classified as severe and moderate classes of the pathology. With each pair, 100% successful classification was achieved. The experiments using the top and right side of head image pair achieved this success rate most frequently. Best results were obtained using the Norm-RGB-Lum color normalization method and the SoftMax data normalization method. These results indicate that there is a difference between the moderate and severe classes with the clinical images. These findings support previous results.

The second set of experiments used the nonclinical images that were classified as severe, moderate and mild classes of the pathology. The best success rate of 83% was achieved with the top and left side image pair. These results are inconclusive but indicate that there may be a difference between the mild, moderate and severe classes in the nonclinical images.

After the first two sets of experiments were performed, comparisons of success rates were made between results on the clinical and nonclinical canine images. Comparing the results we find that the separate classes in the clinical (moderate and severe) images are more homogeneous than the classes (mild, moderate and severe) in the nonclinical images. These findings support previous results.

Introduction and Overview

Thermographic images of canines were analyzed using feature extraction and pattern classification tools. The purpose of this research is to examine the thermographic images of canines of the breed Cavalier King Charles Spaniel to investigate the Chiari malformation, or COMS, pathology.

The primary question here was to explore the use of pairs of images, instead of single images as in previous experiments.

1. Can we successfully differentiate between the classes moderate and severe in the clinical images? Here four different experiments were performed:

➢ Use all top (A1D) and left side (A1LL) head images

➢ Use all top (A1D) and right side (A1LR) head images

➢ Use all front (A1) and left side (A1LL) head images

➢ Use all front (A1) and right side (A1LR) head images

2. Can we successfully differentiate between the classes mild, moderate and severe in the nonclinical images?

➢ Use all top (A1D) and left side (A1LL) head images

➢ Use all top (A1D) and right side (A1LR) head images

➢ Use all front (A1) and left side (A1LL) head images

➢ Use all front (A1) and right side (A1LR) head images

Materials and Methods

Images

The thermal images are from Long Island Veterinary Specialists taken with a Meditherm Med2000 IRIS. The images are in TIF file format as RGB images with 319 columns by 238 rows, 8-bits per pixel per color band. Note that a total of 18 colors are used in these images. The images are thermal images of 34 canines of the breed Cavalier King Charles Spaniel. Included in the data set are images of front of head, top of head, left and right side of head.

Software

The CVIPtools [1] (Computer Vision and Image Processing Tools) software was used to investigate and analyze the images. The primary tools used include Analysis->Features and Analysis->Pattern Classification. The features considered include (see Figure 1) spectral (Fourier), histogram, and texture features. The histogram features include mean, standard deviation, skew, energy and entropy; and the texture features include energy, inertia, correlation, inverse difference, and entropy. The pattern classification tools utilized include various normalization methods, distance and similarity measures and three classification methods (see Figure 2).

[pic]

Figure 1 – CVIPtools Feature Extraction window

[pic]

Figure 2 – CVIPtools Pattern Classification window

In the course of the image analysis it was determined that the previously developed special purpose software would be further enhanced for this phase of the research. The software is designed to ease the analysis process and enable us to perform a significantly greater number of experiments. We added the capability to run groups of experiments simultaneously, with the same parameters, which allowed for the analysis of comparative results. One of the primary factors under investigation was color normalization, and the software was employed to experiment with a number of normalization techniques and methods.

Data Normalization

Data normalization is used to compensate for potential bias due to the varying range on different components of the feature vectors. For example, one component may only range from 1 to 5 and another may range from 1 to 5,000, so a difference of 5 would be a maximum for the first feature, while this same difference might be insignificant for the second feature. Here we explored five data normalization methods:

• Min-max normalization: Maps the data to a specified range, but still retains the relationship between the values

[pic]

• Softmax scaling: A nonlinear method for skewed data distributions, that is, not evenly distributed about the mean. It compresses the data into the range of 0 to 1. The process compresses the data exponentially as it gets farther away from the mean.

[pic]

With the mean and standard deviation defined as in [3].

Color Normalization

For this phase of the research, several different methods for normalizing the colors in the images were compared. Each of the images uses the same color palette consisting of 18 colors. Within an image, each color represents a specific temperature. The color to temperature mapping for an image is based on camera settings that were in effect at the time the image was captured. Since the camera may be recalibrated between each image capture, to obtain best results for each image, within a set of images one color may map to several different temperatures. This shifting of temperature scale between images could introduce noise in the pattern classification process. To compensate for this, a remapping was applied to the color values of each image so that all of the images are mapped to a common temperature scale. Four different color normalization methods were used:

• Luminance: For each pixel in the original image, the color components are weighted as follows to generate a gray level value:

[pic]

• Norm-RGB: Similar to Norm-Gray except that instead of temperatures being mapped to gray levels from 0 to 255, they are mapped to a continuous version of the original color palette. An example of the remapping that occurs when this method is applied to an image for which the temperature range is 21° C to 34° C is shown in Figure 3.

[pic]

Figure 3 – Norm-RGB remapping for image with temperature range of 21° C to 34° C

• Norm-RGB-Lum: This method produces a gray-scale image by applying the Norm-RGB normalization method to an image and then applying the luminance normalization method to the result.

Figures 5a through 5c show an original image, along with the images resulting from each of the normalization methods used:

[pic]

Figure 5a – Original Image

[pic]

Figure 5b – Luminance Normalized Image

[pic]

Figure 5c – Norm-RGB-Lum Normalized Image

Experimental Results and Discussion

Results from Experiment Set #1

Images: Clinical group, image pairs: 1) top and left side, 2) top and right side, 3) front and left side, 4) front and right side.

Classes: moderate and severe.

[pic]

*Note: The number “xx / 262,143” refers to the experiments with a specific feature set, where at least one of the color normalization methods achieved this success rate. In the spring report these results were reported out of 1,310,715, which is the above number (262,143) multiplied by the 5 different color normalization methods.

[pic]

*Note: The number “xx / 262,143” refers to the experiments with a specific feature set, where at least one of the color normalization methods achieved this success rate. In the spring report these results were reported out of 1,310,715, which is the above number (262,143) multiplied by the 5 different color normalization methods.

➢ We believe that the nuber of experiments with 100% classification success indicates that the top of head is the most useful for differentiation of the two classes. These findings support previous results.

➢ These results indicate that there is a difference between the moderate and severe classes in the clinical images.

➢ The results from using pairs of images indicate that the classes are easier to differentiate with the top and either side of head than with the front and either side of head images.

➢ The best results were obtained with the top and right side.

➢ Norm-RGB-Lum provided the best color normalization results. These findings support previous results.

➢ Comparison of the the two above tables indicates that the SoftMax data normalization method is superior to the MinMax method.

➢ The results are limited by the small size of the data sets

Results from Experiment Set #2

Images: Nonclinical group, image pairs: 1) top and left side, 2) top and right side, 3) front and left side, 4) front and right side.

Classes: mild, moderate and severe.

[pic]

*Note: The number “xx / 262,143” refers to the experiments with a specific feature set, where at least one of the color normalization methods achieved this success rate. In the spring report these results were reported out of 1,310,715, which is the above number (262,143) multiplied by the 5 different color normalization methods.

[pic]

*Note: The number “xx / 262,143” refers to the experiments with a specific feature set, where at least one of the color normalization methods achieved this success rate. In the spring report these results were reported out of 1,310,715, which is the above number (262,143) multiplied by the 5 different color normalization methods.

➢ These results indicate that the top and left side of head image pair is best suited for classification. The best success rate of 83% was achieved using this pair.

➢ These results indicate SoftMax data normalization is marginally better than MinMax.

➢ These results are inconclusive but indicate that there may be a difference between the mild, moderate and severe classes in the nonclinical images. These findings support previous results.

Conclusions

Two sets of experiments were performed for a total for of 20,971,440 experiments. Each experiment used a pair of images, instead of a single image as was done in previous experiments. The image pairs used were: 1) top and left side of head, 2) top and right side of head, 3) front and left side of head and 4) front and right side of head. Here we found that the SoftMax data normalization method provided the best results. Regarding color normalization, the Norm-RGB-Lum method provided the best results.

The first set of experiements was to investigate classification of the moderate and severe classes with the clinical images. For these experiments, a maximum success rate of 100% correct classification was achieved. This success rate occurred most frequently in the experiments using the top and right side of head image pair. The Norm-RGB-Lum color normalization and the SoftMax data normalization methods provided the best results. The results indicate that there is a difference between the moderate and severe classes with the clinical images. These findings support previous results.

The second set of experiments used the nonclinical images that were classified as severe, moderate and mild classes of the pathology. The best success rate of 83% was achieved with the top and left side image pair. These results are inconclusive but indicate that there may be a difference between the mild, moderate and severe classes in the nonclinical images.

After the first two sets of experiments were performed, comparisons of success rates were made between results on the clinical and nonclinical canine images. Comparing the results we find that the separate classes in the clinical (moderate and severe) images are more homogeneous than the classes (mild, moderate and severe) in the nonclinical images. This finding supports previous results.

Future Work

• Collect more images for differentiation of the classes of both clinical and nonclinical images, and get the updated classifications

• Observe first hand the image acquisition process to explore standardization of this process and observation of the methods

• Consult the veterinarians to learn more regarding analysis and diagnostic uses of the thermographic images

• Confer with the veterinarians and/or technicians in the development and use of the application-specific software

• Further analyze the results from the 20,971,440 experiments, including analysis of frequency of features relative to classification success and feature set size relative to success

• Investigate use of support vector machines for analysis of veterinary thermographic images

• Research the use of neural networks for the analysis of veterinary thermographic images

• Investigate use of genetic algorithms for the analysis of veterinary thermographic images

• Research methods to determine the “best” feature set(s)

• Enhance the graphical user interface (GUI) for the automatic pattern classification and application-specific software

• Explore finding of feature vector “norms” for various pathologies/abnormalities

• Explore multimodal image analysis

• Investigate development of automatic preprocessing software for thermographic images

• Investigate development of automatic segmentation software for thermographic images

References

1. Veterinary Thermographic Image Analysis, Project Number 7-64878, Report Number 4878-2, August 31, 2007.

2. Veterinary Thermographic Image Analysis, Project Number 7-64878, Report Number 4878-3, January 23, 2008.

3. Veterinary Thermographic Image Analysis, Project Number 7-64878, Report Number 4878-4, June 30, 2008.

4. Computer Imaging: Digital Image Analysis and Processing , Scott E Umbaugh, The CRC Press, Boca Raton, FL, January 2005

5. Pattern Recognition Engineering, Morton Adler and Eric Smith, John-Wiley & Sons, NY, 1993

6. Pattern Classification, 2nd Edition, Richard Duda, Peter Hart, David Stork, John-Wiley & Sons, NY, 2001

7. Pattern Recognition, Sergios Theodoris and Konstantinos Koutroumbas, Academic Press, NY, 2006

Appendix – Result File Names

The experiment result files are not included here (due to their large size), but are in text files with the following file names.

Result Data Files

|Experiment |File name |

|Experiment Set #1 |Exp1-A1-A1LL-MinMax-results.txt |

| |Exp1-A1-A1LR-MinMax-results.txt |

| |Exp1-A1D-A1LL-MinMax-results.txt |

| |Exp1-A1D-A1LR-MinMax-results.txt |

| |Exp1-A1-A1LL-SoftMax-results.txt |

| |Exp1-A1-A1LR-SoftMax-results.txt |

| |Exp1-A1D-A1LL-SoftMax-results.txt |

| |Exp1-A1D-A1LR-SoftMax-results.txt |

|Experiment Set #2 |Exp2-A1-A1LL-MinMax-results.txt |

| |Exp2-A1-A1LR-MinMax-results.txt |

| |Exp2-A1D-A1LL-MinMax-results.txt |

| |Exp2-A1D-A1LR-MinMax-results.txt |

| |Exp2-A1-A1LL-SoftMax-results.txt |

| |Exp2-A1-A1LR-SoftMax-results.txt |

| |Exp2-A1D-A1LL-SoftMax-results.txt |

| |Exp2-A1D-A1LR-SoftMax-results.txt |

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download