Initial implementation of Otsu's image segmentation algorithm, Canny bug fixes,…

core/pom.xml

@@ -2,7 +2,7 @@
 <!--
   ~ C:/Users/chris/IdeaProjects/myriad/core/pom.xml
   ~
-  ~ Copyright (c) 2017 Emphysic LLC.
+  ~ Copyright (c) 2018 Emphysic LLC.
   ~
   ~ Licensed under the Apache License, Version 2.0 (the "License");
   ~ you may not use this file except in compliance with the License.
@@ -41,7 +41,7 @@
         <dependency>
             <groupId>com.github.haifengl</groupId>
             <artifactId>smile-core</artifactId>
-            <version>1.3.0</version>
+            <version>1.5.0</version>
         </dependency>
         <dependency>
             <groupId>com.github</groupId>
@@ -56,7 +56,7 @@
         <dependency>
             <groupId>com.aparapi</groupId>
             <artifactId>aparapi</artifactId>
-            <version>1.3.4</version>
+            <version>1.4.1</version>
         </dependency>
         <dependency>
             <groupId>com.twelvemonkeys.imageio</groupId>
@@ -66,7 +66,7 @@
         <dependency>
             <groupId>org.apache.mahout</groupId>
             <artifactId>mahout-mr</artifactId>
-            <version>0.10.0</version>
+            <version>0.13.0</version>
         </dependency>
         <dependency>
             <groupId>com.mashape.unirest</groupId>

core/src/main/java/com/emphysic/myriad/core/data/ops/CannyOperation.java

 /*
  * com.emphysic.myriad.core.data.ops.CannyOperation
  *
- * Copyright (c) 2017 Emphysic LLC.
+ * Copyright (c) 2018 Emphysic LLC.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -65,10 +65,9 @@ public class CannyOperation implements DatasetOperation {
      * NONE - do not attempt to automatically calculate
      * MEAN - use the mean of the input data to calculate lower and upper thresholds
      * MEDIAN - use the median of the input data to calculate lower and upper thresholds
-     * OTSU - use Otsu's Method (https://en.wikipedia.org/wiki/Otsu%27s_method ) to calculate lower and upper thresholds
-     * (not yet implemented)
+     * OTSU1D - use Otsu's Method (https://en.wikipedia.org/wiki/Otsu%27s_method ) to calculate lower and upper thresholds
      */
-    public enum AutoThreshold {NONE, MEAN, MEDIAN, OTSU}
+    public enum AutoThreshold {NONE, MEAN, MEDIAN, OTSU1D}
 
     /**
      * Whether and how to automatically calculate thresholds
@@ -127,24 +126,33 @@ public class CannyOperation implements DatasetOperation {
     @Override
     public Dataset run(Dataset input) {
         if (input != null) {
-            Dataset result = new Dataset(input.getWidth(), input.getHeight());
             switch (autoThreshold) {
                 case MEAN:
                     // Set thresholds w. mean
                     // (1-sigma)*[mean value] and set the high threshold to (1+sigma)*[mean value]
-                    Double mean = Math.mean(result.getData());
+                    Double mean = Math.mean(input.getData());
                     setLowerThreshold((1 - sigmaThreshold) * mean);
                     setUpperThreshold((1 + sigmaThreshold) * mean);
                     break;
                 case MEDIAN:
                     // Set thresholds w. median
                     // (1-sigma)*[median value] and (1+sigma)*[median value]
-                    Double median = Math.median(result.getData());
+                    Double median = Math.median(input.getData());
                     setLowerThreshold((1 - sigmaThreshold) * median);
                     setUpperThreshold((1 + sigmaThreshold) * median);
                     break;
-                case OTSU:
-                    // Not yet implemented
+                case OTSU1D:
+                    // Sets upper threshold to Otsu's 1D threshold and
+                    // lower to one half this value
+                    int t = Otsu1dOperation.calcThreshold(input, true);
+                    // Thresholds are calculated w. normalized grayscale data between 0-255
+                    // so we need to scale back to our original input
+                    Double max = Math.max(input.getData());
+                    Double min = Math.min(input.getData());
+                    Double threshold = (t / 255) * (max - min) + min;
+                    setLowerThreshold(threshold / 2);
+                    setUpperThreshold(threshold);
+                    break;
                 default:
                     break;
             }
@@ -160,6 +168,7 @@ public class CannyOperation implements DatasetOperation {
             }
             Dataset roundedAngles = new Dataset(angles, origAngles.getWidth(), origAngles.getHeight());
             Dataset origMags = G.getMagnitudes();
+            Dataset result = new Dataset(input.getWidth(), input.getHeight());
             // Non-Maximum Suppression - thin edges by only keeping the local maximum along the gradient direction
             for (int i=0; i<origMags.getWidth(); i++) {
                 for (int j=0; j<origMags.getHeight(); j++) {

core/src/main/java/com/emphysic/myriad/core/data/ops/Otsu1dOperation.java

+/*
+ * com.emphysic.myriad.core.data.ops.Otsu1dOperation
+ *
+ * Copyright (c) 2018 Emphysic LLC.
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package com.emphysic.myriad.core.data.ops;
+
+import com.emphysic.myriad.core.data.io.Dataset;
+import com.emphysic.myriad.core.data.ops.math.Histogram;
+
+import java.util.Map;
+
+/**
+ * Otsu's global thresholding algorithm (1D).  Based on Balarini and Nesmachnow,
+ * "A C++ Implementation of Otsu’s Image Segmentation Method, Image Processing On Line, 6 (2016), pp. 155–164"
+ * https://doi.org/10.5201/ipol.2016.158
+ */
+public class Otsu1dOperation implements DatasetOperation {
+    private static final long serialVersionUID = 1L; // try never to change - indicates backwards compatibility is broken
+    private static final int VERSION = 1; // current implementation version
+
+    @Override
+    public Dataset run(Dataset input) {
+        if (input != null && input.getSize() > 0) {
+            Dataset copy = Histogram.grayScale(input);
+            Dataset result = new Dataset(copy.getWidth(), copy.getHeight());
+            int threshold = calcThreshold(copy);
+            for (int i=0; i<copy.getWidth(); i++) {
+                for (int j=0; j<copy.getHeight(); j++) {
+                    double val = copy.get(i, j);
+                    double out = 0;
+                    if (val > threshold) {
+                        out = 1;
+                    }
+                    result.set(i, j, out);
+                }
+            }
+            return result;
+        }
+        return null;
+    }
+
+    @Override
+    public long getSerializationVersion() {
+        return serialVersionUID;
+    }
+
+    @Override
+    public int getVersion() {
+        return VERSION;
+    }
+
+    @Override
+    public void initCurrentVersion(Map<String, Object> objectMap) {}
+
+    /**
+     * Calculates Otsu's threshold for a given Dataset.
+     * @param input Dataset to threshold
+     * @return threshold value between 0 and 255
+     */
+    public static int calcThreshold(Dataset input) {
+        return calcThreshold(input, false);
+    }
+
+    /**
+     * Calculates Otsu's threshold for a given Dataset.
+     * @param input Dataset to threshold
+     * @param normalize if true, convert input data to gray scale prior to calculating the threshold
+     * @return threshold value between 0 and 255
+     */
+    public static int calcThreshold(Dataset input, boolean normalize) {
+        int threshold = 0;
+        int[] hist = Histogram.hist1d(input, normalize);
+        int suma = 0;
+        for (int i=0; i<hist.length; i++) {
+            suma += i * hist[i];
+        }
+        int q1 = 0;
+        int q2;
+        int sumb = 0;
+        double u1;
+        double u2;
+        double sigma;
+        double var_max = 0;
+        for (int j=0; j<hist.length; j++) {
+            q1 += hist[j];
+            if (q1 == 0) {
+                continue;
+            }
+            q2 = input.getSize() - q1;
+            if (q2 == 0) {
+                break;
+            }
+            sumb += j * hist[j];
+            u1 = sumb / q1;
+            u2 = (suma - sumb) / q2;
+            sigma = q1 * q2 * Math.pow(u1 - u2, 2);
+            if (sigma > var_max) {
+                threshold = j;
+                var_max = sigma;
+            }
+        }
+        return threshold;
+    }
+}

core/src/main/java/com/emphysic/myriad/core/data/ops/math/Histogram.java

+/*
+ * com.emphysic.myriad.core.data.ops.math.Histogram
+ *
+ * Copyright (c) 2018 Emphysic LLC.
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package com.emphysic.myriad.core.data.ops.math;
+
+import com.emphysic.myriad.core.data.io.Dataset;
+import com.emphysic.myriad.core.data.ops.NormalizeSignalOperation;
+
+/**
+ * Simple histogram functions.
+ */
+public class Histogram {
+    /**
+     * Calculates the one-dimensional histogram for a given Dataset.
+     * @param input Dataset to examine
+     * @param normalize if true, input is normalized prior to calculating the histogram
+     * @param max_intensity maximum intensity to use
+     * @return array of size [max_intensity + 1] with a count of the number of elements in input with that value
+     */
+    public static int[] hist1d(Dataset input, boolean normalize, int max_intensity) {
+        Dataset normalized;
+        if (normalize) {
+            normalized = grayScale(input, max_intensity);
+        } else {
+            normalized = new Dataset(input);
+        }
+        int[] hist = new int[max_intensity + 1];
+
+        for (int i=0; i<normalized.getSize(); i++) {
+            int val = (int)normalized.getData()[i];
+            hist[val] += 1;
+        }
+        return hist;
+    }
+
+    /**
+     * Converts a Dataset into a gray scale representation by normalizing between 0 and a maximum value.
+     * @param input input to normalize
+     * @param max_intensity maximum value to use
+     * @return Dataset with scaled amplitudes
+     */
+    public static Dataset grayScale(Dataset input, int max_intensity) {
+        Dataset normalized = new NormalizeSignalOperation().run(input);
+        for(int i=0; i < normalized.getWidth(); i++) {
+            for (int j = 0; j < normalized.getHeight(); j++) {
+                normalized.set(i, j, max_intensity * normalized.get(i, j));
+            }
+        }
+        return normalized;
+    }
+
+    /**
+     * Converts a Dataset into a gray scale representation by normalizing between 0 and 255.
+     * @param input input to normalize
+     * @return Dataset with values between 0 and 255.
+     */
+    public static Dataset grayScale(Dataset input) {
+        return grayScale(input, 255);
+    }
+
+    /**
+     * Calculates the histogram for a given Dataset.
+     * @param input Dataset to examine
+     * @return 255-level histogram
+     */
+    public static int[] hist1d(Dataset input) {
+        return hist1d(input, true, 255);
+    }
+
+    /**
+     * Calculates the histogram for a given Dataset.
+     * @param input Dataset to examine
+     * @param normalize if true, normalize the data between 0-255 prior to calculating the histogram
+     * @return histogram of data
+     */
+    public static int[] hist1d(Dataset input, boolean normalize) {
+        return hist1d(input, normalize, 255);
+    }
+}

network/pom.xml

@@ -2,7 +2,7 @@
 <!--
   ~ C:/Users/chris/IdeaProjects/myriad/network/pom.xml
   ~
-  ~ Copyright (c) 2017 Emphysic LLC.
+  ~ Copyright (c) 2018 Emphysic LLC.
   ~
   ~ Licensed under the Apache License, Version 2.0 (the "License");
   ~ you may not use this file except in compliance with the License.
@@ -32,67 +32,67 @@
         <dependency>
             <groupId>com.typesafe.akka</groupId>
             <artifactId>akka-actor_2.11</artifactId>
-            <version>2.4.17</version>
+            <version>2.5.8</version>
         </dependency>
 
         <dependency>
             <groupId>com.typesafe.akka</groupId>
             <artifactId>akka-agent_2.11</artifactId>
-            <version>2.4.17</version>
+            <version>2.5.8</version>
         </dependency>
 
         <dependency>
             <groupId>com.typesafe.akka</groupId>
             <artifactId>akka-persistence_2.11</artifactId>
-            <version>2.4.17</version>
+            <version>2.5.8</version>
         </dependency>
 
         <dependency>
             <groupId>com.typesafe.akka</groupId>
             <artifactId>akka-persistence-tck_2.11</artifactId>
-            <version>2.4.17</version>
+            <version>2.5.8</version>
         </dependency>
 
         <dependency>
             <groupId>com.typesafe.akka</groupId>
             <artifactId>akka-slf4j_2.11</artifactId>
-            <version>2.4.17</version>
+            <version>2.5.8</version>
         </dependency>
 
         <dependency>
             <groupId>com.typesafe.akka</groupId>
             <artifactId>akka-testkit_2.11</artifactId>
-            <version>2.4.17</version>
+            <version>2.5.8</version>
         </dependency>
 
         <dependency>
             <groupId>com.typesafe.akka</groupId>
             <artifactId>akka-cluster_2.11</artifactId>
-            <version>2.4.17</version>
+            <version>2.5.8</version>
         </dependency>
 
         <dependency>
             <groupId>com.typesafe.akka</groupId>
             <artifactId>akka-cluster-metrics_2.11</artifactId>
-            <version>2.4.17</version>
+            <version>2.5.8</version>
         </dependency>
 
         <dependency>
             <groupId>com.typesafe.akka</groupId>
             <artifactId>akka-cluster-sharding_2.11</artifactId>
-            <version>2.4.17</version>
+            <version>2.5.8</version>
         </dependency>
 
         <dependency>
             <groupId>com.typesafe.akka</groupId>
             <artifactId>akka-cluster-tools_2.11</artifactId>
-            <version>2.4.17</version>
+            <version>2.5.8</version>
         </dependency>
 
         <dependency>
             <groupId>com.typesafe.akka</groupId>
             <artifactId>akka-remote_2.11</artifactId>
-            <version>2.4.17</version>
+            <version>2.5.8</version>
         </dependency>
         <dependency>
             <groupId>com.emphysic.myriad</groupId>