diff --git a/README.md b/README.md
index 000bd21..923105a 100644
--- a/README.md
+++ b/README.md
@@ -1,14 +1,20 @@
 # Image Tools
-Some self used batch image process tools
+Batch image process tools mainly designed for dataset preprocessing
 
 ## Current tools
 
+### Global args
+
+- `thread`  The thread pool size for parallel operations
+
 ### Common
 
+- `crop-rectangle`            Crop a rectangle region of the image
+- `normalize`                 Normalize image to given range
 - `map-color`                 Map one RGB color to another in a given PNG image file
-- `map-color-dir`             Map one RGB color to another for all PNG images in a given folder
+- `map-background-color`      Map all the non-valid color in the image to a given color
 - `split-images`              Split large images to small pieces for augmentation purposes
-- `split-images-with-bias`    Split large images to small pieces with bias for augmentation purposes
+- `split-images-with-bias`    Split large images to small pieces for augmentation purposes with bias (Bias is added between each split)
 - `split-images-with-filter`  Split large images to small pieces with a filter for enough valid pixels
 - `class2rgb`                 Map 8 bit grayscale PNG class image to RGB image
 - `rgb2class`                 Map RGB image to 8 bit grayscale PNG class image
@@ -17,14 +23,13 @@ Some self used batch image process tools
 - `split-dataset`             Split dataset into train and test sets
 - `count-classes`             Count class for 8 bit PNG image & Calc class balance weight
 - `strip-image-edge`          Strip image edges
+- `stich-images`              Stich the splited images back together
 - `calc-mean-std`             Calc the mean and std of a dataset for normalization
+- `calc-iou`                  Calc the IoU of two images
 
 
 ### Yolo
 
 - `split-dataset`  Split dataset into train and test sets Will store result in TXT file
 - `count-types`    Count the object number of each type in the dataset
-- `rgb2yolo`       Convert RGB labels to YOLO TXT format
-
-## TODO
-- [x] Fix hole issue (maybe change from imageproc to opencv contour detection)
\ No newline at end of file
+- `rgb2yolo`       Convert RGB labels to YOLO TXT format
\ No newline at end of file
diff --git a/src/main.rs b/src/main.rs
index bd962f0..4a0f27e 100644
--- a/src/main.rs
+++ b/src/main.rs
@@ -77,6 +77,7 @@ enum CommonCommands {
         dataset_path: String,
     },
 
+    /// Map all the non-valid color in the image to a given color
     MapBackgroundColor {
         #[arg(short, long, help = "In R1,G1,B1;R2,G2,B2 format")]
         valid_colors: String,
@@ -101,7 +102,7 @@ enum CommonCommands {
     },
 
     /// Split large images to small pieces for augmentation purposes with bias
-    /// Bias is added between each split
+    /// (Bias is added between each split)
     SplitImagesWithBias {
         #[arg(short, long, help = "The path for the folder containing images")]
         dataset_path: String,