# Apply Mask This function block filters an image using a mask image (black/white or grayscale) to keep or remove specific parts of the input image. It is intended for simple segmentation and masking workflows in AugeLab Studio. ## đŸ“Ĩ Inputs * `Image` — Main image to be filtered (input socket). * `Mask` — Mask image used to define preserved areas (input socket). Accepts a binary, grayscale or color image. The block will treat white/bright mask pixels as kept areas and dark pixels as removed areas. ## 📤 Outputs * `Filtered` — Resulting image after applying the mask (output socket). ## 🕹ī¸ Controls This block has no extra controls or widget settings. It performs masking based on the two provided inputs. ## 🎨 Features * Works with color (RGB) and grayscale images. * Accepts masks in grayscale or color; the mask is interpreted as a keep/remove map. * Keeps image pixels where mask is bright/white and hides pixels where mask is dark/black. * Designed for fast, visual filtering in processing chains. ## ⚙ī¸ Running mechanism When evaluated, the block checks both inputs, ensures the mask can be used as a binary selector, and then applies the mask to the main image. The output is the original image with non-masked areas removed (or set to black) according to the mask. ## 📝 Usage instructions 1. Provide the image you want to filter to the `Image` input. 2. Provide a mask image to the `Mask` input. Bright/white regions in the mask will be preserved in the output; dark/black regions will be removed. 3. Connect the `Filtered` output to downstream blocks (for visualization, saving, or further processing). Tips: * If the mask and the image differ in size, resize or crop them so they match before connecting. * If your mask is a rough segmentation, combine this block with morphological or smoothing steps to refine results. ## 💡 Tips and Tricks * Create masks from intensity values using the `Image Threshold` or `Image Adaptive Threshold` blocks before feeding them into this block. * Reduce noise in the mask source using `Blur` or `Denoising` to avoid small unwanted holes in the result. * Improve foreground/background separation with `HSV Filter` when color is a strong cue, then convert that result into a mask for this block. * For higher-quality subject extraction, try `Background Removal (RMBG-1.4)` or `Background Removal (BiRefNet)` and feed the resulting mask or cutout into this block. * Use `Morphological Transformations` to clean up small mask artifacts (erode/dilate) before applying the mask. * Crop or focus processing with `Image ROI Select` or `Image ROI` so the mask and image align on the area of interest. * Preview results with `Show Image` and save successful outputs using `Image Logger` or `Image Write`. ## 🛠ī¸ Troubleshooting * Mask appears to do nothing: Verify the `Mask` is aligned with the `Image` and that bright areas correspond to the regions you expect to keep. If sizes differ, use `Image Resize` or crop with `Image ROI Select`. * Output contains speckles or holes: Apply `Blur` or `Morphological Transformations` to the mask to remove small defects. * No visible change after masking: Check that the mask actually contains non-black regions. Use `Image Threshold` to explicitly binarize uncertain masks. * Need to inspect intermediate results: Insert `Show Image` after mask creation steps to verify mask correctness before applying it. ## 📊 Evaluation On execution this block outputs the `Filtered` image. Use the output to visualize, save, or feed into later analysis blocks such as detection or measurement.