Warning, /documentation/digikam-doc/maintenance_tools/maintenance_quality.rst is written in an unsupported language. File is not indexed.
0001 .. meta:: 0002 :description: digiKam Maintenance Tool to Sort Image by Quality 0003 :keywords: digiKam, documentation, user manual, photo management, open source, free, learn, easy, quality, pick-label, deep-learning 0004 0005 .. metadata-placeholder 0006 0007 :authors: - digiKam Team 0008 0009 :license: see Credits and License page for details (https://docs.digikam.org/en/credits_license.html) 0010 0011 .. _maintenance_quality: 0012 0013 Image Quality Sorter 0014 ==================== 0015 0016 .. contents:: 0017 0018 .. figure:: images/maintenance_quality_sorter.webp 0019 :alt: 0020 :align: center 0021 0022 The digiKam Maintenance Options for Image Quality Sorter 0023 0024 The tool will parse the items to determine the **Quality** and sort them with the **Pick Label** properties in database as: 0025 0026 - The **Red** flag for **Rejected**. 0027 0028 - The **Yellow** flag for **Pending**. 0029 0030 - The **Green** flag for **Accepted**. 0031 0032 The Image Quality Sorter assigns Pick Labels to items according to the default settings in :menuselection:`Settings --> Configure digiKam...` and the **Image Quality Sorter** settings panel, or a dedicated settings tuned in this view. 0033 0034 The quality **score** to determine the **Pick Label** property can be evaluated by two ways: using four basic factors sabotaging the images (**Blur**, **Noise**, **Exposure**, and **Compression**), or using a **deep learning neural network engine**. 0035 0036 The first one helps to determine whether images are distorted by the basic factors, however it demands some drawbacks as fine-tuning from the user’s side and it cannot work along the aesthetic image processing. 0037 0038 The second one uses an artificial intelligence approach based on `this model <https://expertphotography.com/aesthetic-photography/>`_ to predict the score. As deep-learning is an end-to-end solution, it doesn’t require hyper-parameter settings, and make this feature easier to use. 0039 0040 .. note:: 0041 0042 Since it is a time consuming process it's a good idea to check Work on all processor cores and restrict the job to certain albums or tags. 0043 0044 All the **Image Quality Sorter** settings is described in :ref:`the dedicated section <imgqsort_settings>` from **Setup Application** section. 0045 0046 The **Scan Mode** settings allows to configure how the database information for the items from the selected collection will be processed. **Clean all and re-scan** reset all data and scan items from scratch, **Scan non-assigned only** will be faster to process non registered items with **Pick Labels**. 0047 0048 While the image quality sorting process is under progress, notification on the bottom right of main windows will be visible to indicate the amount of items already done. 0049 0050 .. figure:: images/maintenance_quality_process.webp 0051 :alt: 0052 :align: center 0053 0054 The digiKam Image Quality Sorter Process Running in the Background 0055 0056 The quality sorting results will appears in :ref:`the Labels view <labels_view>` from the right sidebar. 0057 0058 .. figure:: images/maintenance_pick_labels.webp 0059 :alt: 0060 :align: center 0061 0062 The digiKam Pick Labels from Left Sidebar 0063 0064 .. important:: 0065 0066 To run properly, the process needs the deep-learning model to download at the first run of digiKam. See :ref:`the Quick Start section <quick_start>` for details.