<?xml version="1.0"?>
<article xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" dtd-version="1.4" xsi:noNamespaceSchemaLocation="https://jats.nlm.nih.gov/archiving/1.4/xsd/JATS-archivearticle1.xsd" xml:lang="ru">
  <front>
    <journal-meta>
      <journal-id journal-id-type="ojs">IE</journal-id>
      <journal-title-group>
        <journal-title xml:lang="ru">&#x421;&#x422;&#x420;&#x41E;&#x418;&#x422;&#x415;&#x41B;&#x42C;&#x41D;&#x42B;&#x415; &#x418; &#x414;&#x41E;&#x420;&#x41E;&#x416;&#x41D;&#x42B;&#x415; &#x41C;&#x410;&#x428;&#x418;&#x41D;&#x42B;</journal-title>
        <trans-title-group xml:lang="en">
          <trans-title>STROITEL'NYE I DOROZHNYE MASHINY</trans-title>
        </trans-title-group>
      </journal-title-group>
      <issn pub-type="ppub">0039-2391</issn>
      <publisher>
        <publisher-name>&#x418;&#x41F; &#x41F;&#x43E;&#x434;&#x43A;&#x43E;&#x43B;&#x437;&#x438;&#x43D; &#x41C;.&#x41C;.</publisher-name>
      </publisher>
      <self-uri xlink:href="https://iereview.ru/index.php/IE"/>
    </journal-meta>
    <article-meta>
      <article-id pub-id-type="publisher-id">288</article-id>
      <article-categories>
        <subj-group xml:lang="ru" subj-group-type="heading">
          <subject>&#x41F;&#x420;&#x418;&#x41A;&#x41B;&#x410;&#x414;&#x41D;&#x42B;&#x415; &#x418;&#x421;&#x421;&#x41B;&#x415;&#x414;&#x41E;&#x412;&#x410;&#x41D;&#x418;&#x42F;</subject>
        </subj-group>
      </article-categories>
      <title-group>
        <article-title xml:lang="ru">&#x41F;&#x440;&#x438;&#x43C;&#x435;&#x43D;&#x435;&#x43D;&#x438;&#x435; &#x43E;&#x431;&#x43B;&#x435;&#x433;&#x447;&#x435;&#x43D;&#x43D;&#x44B;&#x445; &#x43D;&#x435;&#x439;&#x440;&#x43E;&#x43D;&#x43D;&#x44B;&#x445; &#x441;&#x435;&#x442;&#x435;&#x439; &#x434;&#x43B;&#x44F; &#x443;&#x43B;&#x443;&#x447;&#x448;&#x435;&#x43D;&#x438;&#x44F; &#x438;&#x437;&#x43E;&#x431;&#x440;&#x430;&#x436;&#x435;&#x43D;&#x438;&#x439; &#x432; &#x440;&#x435;&#x430;&#x43B;&#x44C;&#x43D;&#x43E;&#x43C; &#x432;&#x440;&#x435;&#x43C;&#x435;&#x43D;&#x438; &#x43D;&#x430; &#x43C;&#x43E;&#x431;&#x438;&#x43B;&#x44C;&#x43D;&#x44B;&#x445; &#x443;&#x441;&#x442;&#x440;&#x43E;&#x439;&#x441;&#x442;&#x432;&#x430;&#x445;</article-title>
        <trans-title-group xml:lang="en">
          <trans-title>Application of lightweight neural networks for real-time image enhancement on mobile devices</trans-title>
        </trans-title-group>
      </title-group>
      <contrib-group content-type="author">
        <contrib contrib-type="author">
          <contrib-id contrib-id-type="orcid" authenticated="false">https://orcid.org/0009-0009-2995-9091</contrib-id>
          <name-alternatives>
            <name name-style="western" specific-use="primary" xml:lang="ru">
              <surname>&#x421;&#x44E;&#x439;</surname>
              <given-names>&#x418;&#x444;&#x430;&#x43D;&#x44C;</given-names>
            </name>
            <name name-style="western" xml:lang="en">
              <surname>Xu</surname>
              <given-names>Yifan</given-names>
            </name>
          </name-alternatives>
          <xref ref-type="aff" rid="aff-1"/>
          <email>yfx1634159614@gmail.com</email>
        </contrib>
        <contrib contrib-type="author">
          <contrib-id contrib-id-type="orcid" authenticated="false">https://orcid.org/0009-0001-6277-7187</contrib-id>
          <name-alternatives>
            <name name-style="western" specific-use="primary" xml:lang="ru">
              <surname>&#x424;&#x430;&#x43D;&#x44C;</surname>
              <given-names>&#x41B;&#x438;&#x43D;&#x44C;&#x434;&#x430;</given-names>
            </name>
            <name name-style="western" xml:lang="en">
              <surname>Fan</surname>
              <given-names>Linda</given-names>
            </name>
          </name-alternatives>
          <xref ref-type="aff" rid="aff-1"/>
          <email>fld420605545@gmail.com</email>
        </contrib>
        <contrib contrib-type="author">
          <contrib-id contrib-id-type="orcid" authenticated="false">https://orcid.org/0009-0001-7560-842X</contrib-id>
          <name-alternatives>
            <name name-style="western" specific-use="primary" xml:lang="ru">
              <surname>&#x421;&#x44F;</surname>
              <given-names>&#x42D;&#x43D;&#x44C;&#x434;&#x43E;</given-names>
            </name>
            <name name-style="western" xml:lang="en">
              <surname>Xia</surname>
              <given-names>Enduo</given-names>
            </name>
          </name-alternatives>
          <xref ref-type="aff" rid="aff-1"/>
          <email>enduoxia@gmail.com</email>
        </contrib>
        <contrib contrib-type="author">
          <contrib-id contrib-id-type="orcid" authenticated="false">https://orcid.org/0009-0001-5319-3374</contrib-id>
          <name-alternatives>
            <name name-style="western" specific-use="primary" xml:lang="ru">
              <surname>&#x425;&#x44D;</surname>
              <given-names>&#x425;&#x443;&#x43D;&#x44C;&#x44F;&#x43D;&#x44C;</given-names>
            </name>
            <name name-style="western" xml:lang="en">
              <surname>He</surname>
              <given-names>Hongyan</given-names>
            </name>
          </name-alternatives>
          <xref ref-type="aff" rid="aff-1"/>
          <email>hehy9527@gmail.com</email>
        </contrib>
      </contrib-group>
      <aff-alternatives id="aff-1">
        <aff xml:lang="ru">
          <institution content-type="orgname">&#x411;&#x435;&#x43B;&#x43E;&#x440;&#x443;&#x441;&#x441;&#x43A;&#x438;&#x439; &#x433;&#x43E;&#x441;&#x443;&#x434;&#x430;&#x440;&#x441;&#x442;&#x432;&#x435;&#x43D;&#x43D;&#x44B;&#x439; &#x443;&#x43D;&#x438;&#x432;&#x435;&#x440;&#x441;&#x438;&#x442;&#x435;&#x442;, 220030, &#x420;&#x435;&#x441;&#x43F;&#x443;&#x431;&#x43B;&#x438;&#x43A;&#x430; &#x411;&#x435;&#x43B;&#x430;&#x440;&#x443;&#x441;&#x44C;, &#x433;. &#x41C;&#x438;&#x43D;&#x441;&#x43A;, &#x43F;&#x440;. &#x41D;&#x435;&#x437;&#x430;&#x432;&#x438;&#x441;&#x438;&#x43C;&#x43E;&#x441;&#x442;&#x438;, 4</institution>
        </aff>
        <aff xml:lang="en">
          <institution content-type="orgname">Belarusian State University, 4 Independence Avenue, Minsk, 220030, Belarus</institution>
        </aff>
      </aff-alternatives>
      <pub-date date-type="pub" publication-format="epub">
        <day>30</day>
        <month>03</month>
        <year>2026</year>
      </pub-date>
      <volume>70</volume>
      <issue>3</issue>
      <fpage>250</fpage>
      <lpage>258</lpage>
      <permissions>
        <copyright-statement xml:lang="ru">&#xA9; 2026 &#x421;&#x422;&#x420;&#x41E;&#x418;&#x422;&#x415;&#x41B;&#x42C;&#x41D;&#x42B;&#x415; &#x418; &#x414;&#x41E;&#x420;&#x41E;&#x416;&#x41D;&#x42B;&#x415; &#x41C;&#x410;&#x428;&#x418;&#x41D;&#x42B;. &#x412;&#x441;&#x435; &#x43F;&#x440;&#x430;&#x432;&#x430; &#x437;&#x430;&#x449;&#x438;&#x449;&#x435;&#x43D;&#x44B;.</copyright-statement>
        <copyright-statement xml:lang="en">&#xA9; 2026 STROITEL'NYE I DOROZHNYE MASHINY. All rights reserved.</copyright-statement>
        <copyright-year>2026</copyright-year>
        <copyright-holder xml:lang="ru">&#x421;&#x422;&#x420;&#x41E;&#x418;&#x422;&#x415;&#x41B;&#x42C;&#x41D;&#x42B;&#x415; &#x418; &#x414;&#x41E;&#x420;&#x41E;&#x416;&#x41D;&#x42B;&#x415; &#x41C;&#x410;&#x428;&#x418;&#x41D;&#x42B;</copyright-holder>
        <copyright-holder xml:lang="en">STROITEL'NYE I DOROZHNYE MASHINY</copyright-holder>
        <license license-type="open-access" specific-use="metadata" xlink:href="https://creativecommons.org/publicdomain/zero/1.0/" xml:lang="ru">
          <license-p>&#x41C;&#x435;&#x442;&#x430;&#x434;&#x430;&#x43D;&#x43D;&#x44B;&#x435; &#x43D;&#x430;&#x441;&#x442;&#x43E;&#x44F;&#x449;&#x435;&#x439; &#x437;&#x430;&#x43F;&#x438;&#x441;&#x438; &#x440;&#x430;&#x441;&#x43F;&#x440;&#x43E;&#x441;&#x442;&#x440;&#x430;&#x43D;&#x44F;&#x44E;&#x442;&#x441;&#x44F; &#x43D;&#x430; &#x443;&#x441;&#x43B;&#x43E;&#x432;&#x438;&#x44F;&#x445; Creative Commons CC0 1.0 (&#x43F;&#x435;&#x440;&#x435;&#x434;&#x430;&#x447;&#x430; &#x432; &#x43E;&#x431;&#x449;&#x435;&#x441;&#x442;&#x432;&#x435;&#x43D;&#x43D;&#x43E;&#x435; &#x434;&#x43E;&#x441;&#x442;&#x43E;&#x44F;&#x43D;&#x438;&#x435;).</license-p>
        </license>
        <license license-type="open-access" specific-use="metadata" xlink:href="https://creativecommons.org/publicdomain/zero/1.0/" xml:lang="en">
          <license-p>The metadata of this record are distributed under the Creative Commons CC0 1.0 Universal Public Domain Dedication.</license-p>
        </license>
      </permissions>
      <self-uri xlink:href="https://iereview.ru/index.php/IE/article/view/288"/>
      <abstract xml:lang="ru">
        <p>&#x421;&#x43C;&#x430;&#x440;&#x442;&#x444;&#x43E;&#x43D;&#x44B; &#x438; &#x43F;&#x43B;&#x430;&#x43D;&#x448;&#x435;&#x442;&#x44B; &#x437;&#x430; &#x43F;&#x43E;&#x441;&#x43B;&#x435;&#x434;&#x43D;&#x435;&#x435; &#x434;&#x435;&#x441;&#x44F;&#x442;&#x438;&#x43B;&#x435;&#x442;&#x438;&#x435; &#x43F;&#x440;&#x435;&#x432;&#x440;&#x430;&#x442;&#x438;&#x43B;&#x438;&#x441;&#x44C; &#x432; &#x43F;&#x43E;&#x43B;&#x43D;&#x43E;&#x446;&#x435;&#x43D;&#x43D;&#x44B;&#x435; &#x432;&#x44B;&#x447;&#x438;&#x441;&#x43B;&#x438;&#x442;&#x435;&#x43B;&#x44C;&#x43D;&#x44B;&#x435; &#x43F;&#x43B;&#x430;&#x442;&#x444;&#x43E;&#x440;&#x43C;&#x44B; &#x2013; &#x438; &#x432;&#x43C;&#x435;&#x441;&#x442;&#x435; &#x441; &#x44D;&#x442;&#x438;&#x43C; &#x441;&#x434;&#x432;&#x438;&#x433;&#x43E;&#x43C; &#x432;&#x43E;&#x437;&#x43D;&#x438;&#x43A; &#x437;&#x430;&#x43A;&#x43E;&#x43D;&#x43E;&#x43C;&#x435;&#x440;&#x43D;&#x44B;&#x439; &#x432;&#x43E;&#x43F;&#x440;&#x43E;&#x441;: &#x43D;&#x430;&#x441;&#x43A;&#x43E;&#x43B;&#x44C;&#x43A;&#x43E; &#x441;&#x43B;&#x43E;&#x436;&#x43D;&#x44B;&#x435; &#x437;&#x430;&#x434;&#x430;&#x447;&#x438; &#x43E;&#x431;&#x440;&#x430;&#x431;&#x43E;&#x442;&#x43A;&#x438; &#x438;&#x437;&#x43E;&#x431;&#x440;&#x430;&#x436;&#x435;&#x43D;&#x438;&#x439; &#x43C;&#x43E;&#x436;&#x43D;&#x43E; &#x440;&#x435;&#x448;&#x430;&#x442;&#x44C; &#x43D;&#x435;&#x43F;&#x43E;&#x441;&#x440;&#x435;&#x434;&#x441;&#x442;&#x432;&#x435;&#x43D;&#x43D;&#x43E; &#x43D;&#x430; &#x443;&#x441;&#x442;&#x440;&#x43E;&#x439;&#x441;&#x442;&#x432;&#x435;, &#x43D;&#x435; &#x43F;&#x440;&#x438;&#x431;&#x435;&#x433;&#x430;&#x44F; &#x43A; &#x443;&#x434;&#x430;&#x43B;&#x435;&#x43D;&#x43D;&#x44B;&#x43C; &#x441;&#x435;&#x440;&#x432;&#x435;&#x440;&#x430;&#x43C;. &#x41D;&#x430;&#x441;&#x442;&#x43E;&#x44F;&#x449;&#x430;&#x44F; &#x441;&#x442;&#x430;&#x442;&#x44C;&#x44F; &#x43F;&#x43E;&#x441;&#x432;&#x44F;&#x449;&#x435;&#x43D;&#x430; &#x43E;&#x431;&#x43B;&#x435;&#x433;&#x447;&#x435;&#x43D;&#x43D;&#x44B;&#x43C; &#x43D;&#x435;&#x439;&#x440;&#x43E;&#x43D;&#x43D;&#x44B;&#x43C; &#x441;&#x435;&#x442;&#x44F;&#x43C; &#x43A;&#x430;&#x43A; &#x438;&#x43D;&#x441;&#x442;&#x440;&#x443;&#x43C;&#x435;&#x43D;&#x442;&#x443; &#x443;&#x43B;&#x443;&#x447;&#x448;&#x435;&#x43D;&#x438;&#x44F; &#x438;&#x437;&#x43E;&#x431;&#x440;&#x430;&#x436;&#x435;&#x43D;&#x438;&#x439; &#x432; &#x440;&#x435;&#x430;&#x43B;&#x44C;&#x43D;&#x43E;&#x43C; &#x432;&#x440;&#x435;&#x43C;&#x435;&#x43D;&#x438; &#x432; &#x443;&#x441;&#x43B;&#x43E;&#x432;&#x438;&#x44F;&#x445; &#x430;&#x43F;&#x43F;&#x430;&#x440;&#x430;&#x442;&#x43D;&#x44B;&#x445; &#x43E;&#x433;&#x440;&#x430;&#x43D;&#x438;&#x447;&#x435;&#x43D;&#x438;&#x439; &#x43C;&#x43E;&#x431;&#x438;&#x43B;&#x44C;&#x43D;&#x44B;&#x445; &#x43F;&#x43B;&#x430;&#x442;&#x444;&#x43E;&#x440;&#x43C;. &#x412; &#x446;&#x435;&#x43D;&#x442;&#x440;&#x435; &#x430;&#x43D;&#x430;&#x43B;&#x438;&#x437;&#x430; &#x2013; &#x430;&#x440;&#x445;&#x438;&#x442;&#x435;&#x43A;&#x442;&#x443;&#x440;&#x43D;&#x44B;&#x435; &#x440;&#x435;&#x448;&#x435;&#x43D;&#x438;&#x44F;, &#x43D;&#x430;&#x43F;&#x440;&#x430;&#x432;&#x43B;&#x435;&#x43D;&#x43D;&#x44B;&#x435; &#x43D;&#x430; &#x441;&#x43E;&#x43A;&#x440;&#x430;&#x449;&#x435;&#x43D;&#x438;&#x435; &#x447;&#x438;&#x441;&#x43B;&#x430; &#x43F;&#x430;&#x440;&#x430;&#x43C;&#x435;&#x442;&#x440;&#x43E;&#x432; &#x43C;&#x43E;&#x434;&#x435;&#x43B;&#x438;, &#x441;&#x43D;&#x438;&#x436;&#x435;&#x43D;&#x438;&#x435; &#x432;&#x44B;&#x447;&#x438;&#x441;&#x43B;&#x438;&#x442;&#x435;&#x43B;&#x44C;&#x43D;&#x43E;&#x439; &#x43D;&#x430;&#x433;&#x440;&#x443;&#x437;&#x43A;&#x438; &#x438; &#x443;&#x43C;&#x435;&#x43D;&#x44C;&#x448;&#x435;&#x43D;&#x438;&#x435; &#x44D;&#x43D;&#x435;&#x440;&#x433;&#x43E;&#x43F;&#x43E;&#x442;&#x440;&#x435;&#x431;&#x43B;&#x435;&#x43D;&#x438;&#x44F; &#x43F;&#x440;&#x438; &#x441;&#x43E;&#x445;&#x440;&#x430;&#x43D;&#x435;&#x43D;&#x438;&#x438; &#x43F;&#x440;&#x438;&#x435;&#x43C;&#x43B;&#x435;&#x43C;&#x43E;&#x433;&#x43E; &#x43A;&#x430;&#x447;&#x435;&#x441;&#x442;&#x432;&#x430; &#x432;&#x44B;&#x445;&#x43E;&#x434;&#x43D;&#x43E;&#x433;&#x43E; &#x438;&#x437;&#x43E;&#x431;&#x440;&#x430;&#x436;&#x435;&#x43D;&#x438;&#x44F;. &#x420;&#x430;&#x441;&#x441;&#x43C;&#x430;&#x442;&#x440;&#x438;&#x432;&#x430;&#x435;&#x43C;&#x44B;&#x435; &#x437;&#x430;&#x434;&#x430;&#x447;&#x438; &#x43E;&#x445;&#x432;&#x430;&#x442;&#x44B;&#x432;&#x430;&#x44E;&#x442; &#x43F;&#x43E;&#x432;&#x44B;&#x448;&#x435;&#x43D;&#x438;&#x435; &#x440;&#x435;&#x437;&#x43A;&#x43E;&#x441;&#x442;&#x438;, &#x43F;&#x43E;&#x434;&#x430;&#x432;&#x43B;&#x435;&#x43D;&#x438;&#x435; &#x448;&#x443;&#x43C;&#x430;, &#x43A;&#x43E;&#x440;&#x440;&#x435;&#x43A;&#x446;&#x438;&#x44E; &#x44F;&#x440;&#x43A;&#x43E;&#x441;&#x442;&#x438; &#x438; &#x43A;&#x43E;&#x43D;&#x442;&#x440;&#x430;&#x441;&#x442;&#x43D;&#x43E;&#x441;&#x442;&#x438;, &#x430; &#x442;&#x430;&#x43A;&#x436;&#x435; &#x43E;&#x431;&#x449;&#x435;&#x435; &#x443;&#x43B;&#x443;&#x447;&#x448;&#x435;&#x43D;&#x438;&#x435; &#x432;&#x438;&#x437;&#x443;&#x430;&#x43B;&#x44C;&#x43D;&#x43E;&#x433;&#x43E; &#x43A;&#x430;&#x447;&#x435;&#x441;&#x442;&#x432;&#x430;. &#x41E;&#x441;&#x43E;&#x431;&#x43E;&#x435; &#x43C;&#x435;&#x441;&#x442;&#x43E; &#x437;&#x430;&#x43D;&#x438;&#x43C;&#x430;&#x44E;&#x442; &#x43C;&#x435;&#x442;&#x43E;&#x434;&#x44B; &#x43E;&#x43F;&#x442;&#x438;&#x43C;&#x438;&#x437;&#x430;&#x446;&#x438;&#x438; &#x43C;&#x43E;&#x434;&#x435;&#x43B;&#x435;&#x439; &#x2013; &#x43A;&#x432;&#x430;&#x43D;&#x442;&#x438;&#x437;&#x430;&#x446;&#x438;&#x44F;, &#x43F;&#x440;&#x43E;&#x440;&#x435;&#x436;&#x438;&#x432;&#x430;&#x43D;&#x438;&#x435; &#x438; &#x441;&#x43D;&#x438;&#x436;&#x435;&#x43D;&#x438;&#x435; &#x440;&#x430;&#x437;&#x43C;&#x435;&#x440;&#x43D;&#x43E;&#x441;&#x442;&#x438; &#x2013; &#x438; &#x441;&#x43F;&#x435;&#x446;&#x438;&#x444;&#x438;&#x43A;&#x430; &#x438;&#x445; &#x43F;&#x440;&#x430;&#x43A;&#x442;&#x438;&#x447;&#x435;&#x441;&#x43A;&#x43E;&#x433;&#x43E; &#x432;&#x43D;&#x435;&#x434;&#x440;&#x435;&#x43D;&#x438;&#x44F; &#x432; &#x43C;&#x43E;&#x431;&#x438;&#x43B;&#x44C;&#x43D;&#x44B;&#x435; &#x43F;&#x440;&#x43E;&#x433;&#x440;&#x430;&#x43C;&#x43C;&#x43D;&#x44B;&#x435; &#x441;&#x440;&#x435;&#x434;&#x44B;. &#x41A;&#x43E;&#x43C;&#x43F;&#x430;&#x43A;&#x442;&#x43D;&#x44B;&#x435; &#x441;&#x432;&#x435;&#x440;&#x442;&#x43E;&#x447;&#x43D;&#x44B;&#x435; &#x430;&#x440;&#x445;&#x438;&#x442;&#x435;&#x43A;&#x442;&#x443;&#x440;&#x44B; &#x440;&#x430;&#x441;&#x441;&#x43C;&#x430;&#x442;&#x440;&#x438;&#x432;&#x430;&#x44E;&#x442;&#x441;&#x44F; &#x43D;&#x435; &#x438;&#x437;&#x43E;&#x43B;&#x438;&#x440;&#x43E;&#x432;&#x430;&#x43D;&#x43D;&#x43E;, &#x430; &#x432; &#x441;&#x432;&#x44F;&#x437;&#x43A;&#x435; &#x441; &#x43A;&#x43E;&#x43D;&#x43A;&#x440;&#x435;&#x442;&#x43D;&#x44B;&#x43C;&#x438; &#x441;&#x446;&#x435;&#x43D;&#x430;&#x440;&#x438;&#x44F;&#x43C;&#x438; &#x43F;&#x440;&#x438;&#x43C;&#x435;&#x43D;&#x435;&#x43D;&#x438;&#x44F;: &#x43C;&#x43E;&#x431;&#x438;&#x43B;&#x44C;&#x43D;&#x43E;&#x439; &#x444;&#x43E;&#x442;&#x43E;&#x433;&#x440;&#x430;&#x444;&#x438;&#x435;&#x439;, &#x434;&#x43E;&#x43F;&#x43E;&#x43B;&#x43D;&#x435;&#x43D;&#x43D;&#x43E;&#x439; &#x440;&#x435;&#x430;&#x43B;&#x44C;&#x43D;&#x43E;&#x441;&#x442;&#x44C;&#x44E;, &#x442;&#x435;&#x43B;&#x435;&#x43C;&#x435;&#x434;&#x438;&#x446;&#x438;&#x43D;&#x43E;&#x439; &#x438; &#x438;&#x43D;&#x442;&#x435;&#x43B;&#x43B;&#x435;&#x43A;&#x442;&#x443;&#x430;&#x43B;&#x44C;&#x43D;&#x44B;&#x43C;&#x438; &#x432;&#x438;&#x437;&#x443;&#x430;&#x43B;&#x44C;&#x43D;&#x44B;&#x43C;&#x438; &#x441;&#x438;&#x441;&#x442;&#x435;&#x43C;&#x430;&#x43C;&#x438;. &#x41F;&#x43E;&#x43B;&#x443;&#x447;&#x435;&#x43D;&#x43D;&#x44B;&#x435; &#x440;&#x435;&#x437;&#x443;&#x43B;&#x44C;&#x442;&#x430;&#x442;&#x44B; &#x441;&#x432;&#x438;&#x434;&#x435;&#x442;&#x435;&#x43B;&#x44C;&#x441;&#x442;&#x432;&#x443;&#x44E;&#x442; &#x43E; &#x442;&#x43E;&#x43C;, &#x447;&#x442;&#x43E; &#x43E;&#x431;&#x43B;&#x435;&#x433;&#x447;&#x435;&#x43D;&#x43D;&#x44B;&#x435; &#x43C;&#x43E;&#x434;&#x435;&#x43B;&#x438; &#x433;&#x43B;&#x443;&#x431;&#x43E;&#x43A;&#x43E;&#x433;&#x43E; &#x43E;&#x431;&#x443;&#x447;&#x435;&#x43D;&#x438;&#x44F; &#x441;&#x43F;&#x43E;&#x441;&#x43E;&#x431;&#x43D;&#x44B; &#x43E;&#x431;&#x435;&#x441;&#x43F;&#x435;&#x447;&#x438;&#x432;&#x430;&#x442;&#x44C; &#x43E;&#x431;&#x440;&#x430;&#x431;&#x43E;&#x442;&#x43A;&#x443; &#x438;&#x437;&#x43E;&#x431;&#x440;&#x430;&#x436;&#x435;&#x43D;&#x438;&#x439; &#x432; &#x440;&#x435;&#x430;&#x43B;&#x44C;&#x43D;&#x43E;&#x43C; &#x432;&#x440;&#x435;&#x43C;&#x435;&#x43D;&#x438; &#x43F;&#x440;&#x438; &#x440;&#x430;&#x437;&#x443;&#x43C;&#x43D;&#x43E;&#x43C; &#x431;&#x430;&#x43B;&#x430;&#x43D;&#x441;&#x435; &#x43C;&#x435;&#x436;&#x434;&#x443; &#x43A;&#x430;&#x447;&#x435;&#x441;&#x442;&#x432;&#x43E;&#x43C;, &#x441;&#x43A;&#x43E;&#x440;&#x43E;&#x441;&#x442;&#x44C;&#x44E; &#x438; &#x440;&#x430;&#x441;&#x445;&#x43E;&#x434;&#x43E;&#x43C; &#x440;&#x435;&#x441;&#x443;&#x440;&#x441;&#x43E;&#x432; &#x443;&#x441;&#x442;&#x440;&#x43E;&#x439;&#x441;&#x442;&#x432;&#x430;. &#x414;&#x430;&#x43B;&#x44C;&#x43D;&#x435;&#x439;&#x448;&#x435;&#x435; &#x440;&#x430;&#x437;&#x432;&#x438;&#x442;&#x438;&#x435; &#x441;&#x43F;&#x435;&#x446;&#x438;&#x430;&#x43B;&#x438;&#x437;&#x438;&#x440;&#x43E;&#x432;&#x430;&#x43D;&#x43D;&#x44B;&#x445; &#x43A;&#x43E;&#x43C;&#x43F;&#x430;&#x43A;&#x442;&#x43D;&#x44B;&#x445; &#x430;&#x440;&#x445;&#x438;&#x442;&#x435;&#x43A;&#x442;&#x443;&#x440;, &#x430;&#x434;&#x430;&#x43F;&#x442;&#x438;&#x440;&#x43E;&#x432;&#x430;&#x43D;&#x43D;&#x44B;&#x445; &#x43A; &#x442;&#x440;&#x435;&#x431;&#x43E;&#x432;&#x430;&#x43D;&#x438;&#x44F;&#x43C; &#x43C;&#x43E;&#x431;&#x438;&#x43B;&#x44C;&#x43D;&#x44B;&#x445; &#x43F;&#x43B;&#x430;&#x442;&#x444;&#x43E;&#x440;&#x43C; &#x438; &#x441;&#x446;&#x435;&#x43D;&#x430;&#x440;&#x438;&#x44F;&#x43C; &#x43B;&#x43E;&#x43A;&#x430;&#x43B;&#x44C;&#x43D;&#x43E;&#x439; &#x43E;&#x431;&#x440;&#x430;&#x431;&#x43E;&#x442;&#x43A;&#x438;, &#x43F;&#x440;&#x435;&#x434;&#x441;&#x442;&#x430;&#x432;&#x43B;&#x44F;&#x435;&#x442;&#x441;&#x44F; &#x43F;&#x435;&#x440;&#x441;&#x43F;&#x435;&#x43A;&#x442;&#x438;&#x432;&#x43D;&#x44B;&#x43C; &#x43D;&#x430;&#x43F;&#x440;&#x430;&#x432;&#x43B;&#x435;&#x43D;&#x438;&#x435;&#x43C; &#x43A;&#x430;&#x43A; &#x441; &#x438;&#x441;&#x441;&#x43B;&#x435;&#x434;&#x43E;&#x432;&#x430;&#x442;&#x435;&#x43B;&#x44C;&#x441;&#x43A;&#x43E;&#x439;, &#x442;&#x430;&#x43A; &#x438; &#x441; &#x43F;&#x440;&#x438;&#x43A;&#x43B;&#x430;&#x434;&#x43D;&#x43E;&#x439; &#x442;&#x43E;&#x447;&#x43A;&#x438; &#x437;&#x440;&#x435;&#x43D;&#x438;&#x44F;.</p>
      </abstract>
      <trans-abstract xml:lang="en">
        <p>Mobile devices have transformed markedly over the past decade &#x2013; their growing computational power has made it both possible and practical to handle image enhancement directly on-device, without routing data through remote servers. This article focuses on lightweight neural networks as a tool for real-time image enhancement under the hardware constraints of mobile platforms. The analysis centers on architectural approaches aimed at reducing parameter counts, lowering computational load, and minimizing energy consumption while maintaining acceptable output image quality. The tasks under consideration include sharpness enhancement, noise suppression, brightness and contrast correction, and general visual quality improvement. Particular attention is given to model optimization techniques &#x2013; quantization, pruning, and dimensionality reduction &#x2013; alongside the practical challenges of deploying such models in mobile software environments. Compact convolutional architectures are examined not in isolation but in relation to concrete application scenarios: mobile photography, augmented reality, telemedicine, and intelligent visual systems. The results indicate that lightweight deep learning models can deliver real-time image processing at a reasonable balance between output quality, processing speed, and device resource consumption. Further development of specialized compact architectures tailored to the demands of mobile platforms and local image processing scenarios is identified as a promising direction for both research and applied work.</p>
      </trans-abstract>
      <kwd-group xml:lang="ru">
        <kwd>&#x43E;&#x431;&#x43B;&#x435;&#x433;&#x447;&#x435;&#x43D;&#x43D;&#x44B;&#x435; &#x43D;&#x435;&#x439;&#x440;&#x43E;&#x43D;&#x43D;&#x44B;&#x435; &#x441;&#x435;&#x442;&#x438;</kwd>
        <kwd>&#x443;&#x43B;&#x443;&#x447;&#x448;&#x435;&#x43D;&#x438;&#x435; &#x438;&#x437;&#x43E;&#x431;&#x440;&#x430;&#x436;&#x435;&#x43D;&#x438;&#x439;</kwd>
        <kwd>&#x43E;&#x431;&#x440;&#x430;&#x431;&#x43E;&#x442;&#x43A;&#x430; &#x438;&#x437;&#x43E;&#x431;&#x440;&#x430;&#x436;&#x435;&#x43D;&#x438;&#x439; &#x432; &#x440;&#x435;&#x430;&#x43B;&#x44C;&#x43D;&#x43E;&#x43C; &#x432;&#x440;&#x435;&#x43C;&#x435;&#x43D;&#x438;</kwd>
        <kwd>&#x43C;&#x43E;&#x431;&#x438;&#x43B;&#x44C;&#x43D;&#x44B;&#x435; &#x443;&#x441;&#x442;&#x440;&#x43E;&#x439;&#x441;&#x442;&#x432;&#x430;</kwd>
        <kwd>&#x433;&#x43B;&#x443;&#x431;&#x43E;&#x43A;&#x43E;&#x435; &#x43E;&#x431;&#x443;&#x447;&#x435;&#x43D;&#x438;&#x435;</kwd>
        <kwd>&#x441;&#x432;&#x435;&#x440;&#x442;&#x43E;&#x447;&#x43D;&#x44B;&#x435; &#x43D;&#x435;&#x439;&#x440;&#x43E;&#x43D;&#x43D;&#x44B;&#x435; &#x441;&#x435;&#x442;&#x438;</kwd>
        <kwd>&#x43E;&#x43F;&#x442;&#x438;&#x43C;&#x438;&#x437;&#x430;&#x446;&#x438;&#x44F; &#x43D;&#x435;&#x439;&#x440;&#x43E;&#x441;&#x435;&#x442;&#x435;&#x432;&#x44B;&#x445; &#x43C;&#x43E;&#x434;&#x435;&#x43B;&#x435;&#x439;</kwd>
        <kwd>&#x43A;&#x43E;&#x43C;&#x43F;&#x44C;&#x44E;&#x442;&#x435;&#x440;&#x43D;&#x43E;&#x435; &#x437;&#x440;&#x435;&#x43D;&#x438;&#x435;</kwd>
      </kwd-group>
      <kwd-group xml:lang="en">
        <kwd>lightweight neural networks</kwd>
        <kwd>image enhancement</kwd>
        <kwd>real-time image processing</kwd>
        <kwd>mobile devices</kwd>
        <kwd>deep learning</kwd>
        <kwd>convolutional neural networks</kwd>
        <kwd>neural network model optimization</kwd>
        <kwd>computer vision</kwd>
      </kwd-group>
      <funding-group>
        <funding-statement xml:lang="ru">&#x418;&#x441;&#x441;&#x43B;&#x435;&#x434;&#x43E;&#x432;&#x430;&#x43D;&#x438;&#x435; &#x432;&#x44B;&#x43F;&#x43E;&#x43B;&#x43D;&#x435;&#x43D;&#x43E; &#x431;&#x435;&#x437; &#x432;&#x43D;&#x435;&#x448;&#x43D;&#x435;&#x433;&#x43E; &#x444;&#x438;&#x43D;&#x430;&#x43D;&#x441;&#x438;&#x440;&#x43E;&#x432;&#x430;&#x43D;&#x438;&#x44F;.</funding-statement>
        <funding-statement xml:lang="en">The study was conducted without external funding.</funding-statement>
      </funding-group>
      <counts>
        <page-count count="9"/>
      </counts>
      <custom-meta-group>
        <custom-meta>
          <meta-name>metadata-license</meta-name>
          <meta-value>
            <ext-link ext-link-type="uri" xlink:href="https://creativecommons.org/publicdomain/zero/1.0/">CC0 1.0</ext-link>
          </meta-value>
        </custom-meta>
      </custom-meta-group>
    </article-meta>
  </front>
  <back>
    <ref-list xml:lang="ru">
      <title>&#x421;&#x43F;&#x438;&#x441;&#x43E;&#x43A; &#x43B;&#x438;&#x442;&#x435;&#x440;&#x430;&#x442;&#x443;&#x440;&#x44B;</title>
      <ref id="R1">
        <mixed-citation>Ahn N., Kang B., Sohn K.-A. Fast, accurate, and lightweight super-resolution with cascading residual network // Computer vision ECCV 2018. 2018. pp. 252-268.</mixed-citation>
      </ref>
      <ref id="R2">
        <mixed-citation>Dong C., Loy C.C., He K., Tang X. Image super-resolution using deep convolutional networks // IEEE Transactions on pattern analysis and machine intelligence. 2016. Vol. 38. &#x2116; 2. pp. 295-307.</mixed-citation>
      </ref>
      <ref id="R3">
        <mixed-citation>Dong C., Loy C.C., Tang X. Accelerating the super-resolution convolutional neural network // Computer vision ECCV 2016. Lecture notes in computer science. 2016. Vol. 9906. pp. 391-407.</mixed-citation>
      </ref>
      <ref id="R4">
        <mixed-citation>Han K., Wang Y., Tian Q., Guo J., Xu C., Xu C. GhostNet: More Features from Cheap Operations // Proceedings of the IEEE/CVF &#x441;onf. on computer vision and pattern recognition. 2020. pp. 1580-1589.</mixed-citation>
      </ref>
      <ref id="R5">
        <mixed-citation>Howard A., Sandler M., Chu G., Chen L.-C., Chen B., Tan M., Wang W., Zhu Y., Pang R., Vasudevan V., Le Q.V., Adam H. Searching for MobileNetV3 // Proceedings of the IEEE/CVF Inter. Conf. on computer vision. 2019. pp. 1314-1324.</mixed-citation>
      </ref>
      <ref id="R6">
        <mixed-citation>Howard A.G., Zhu M., Chen B., Kalenichenko D., Wang W., Weyand T., Andreetto M., Adam H. MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications // arXiv. 2017.</mixed-citation>
      </ref>
      <ref id="R7">
        <mixed-citation>Hui Z., Gao X., Yang Y., Wang X. Lightweight image super-resolution with information multi distillation network // Proceedings of the 27th ACM Inter. &#x441;onf. on multimedia. 2019. pp. 2024-2032.</mixed-citation>
      </ref>
      <ref id="R8">
        <mixed-citation>Ignatov A., Malivenko G., Timofte R., Tseng Y., Xu Y.-S., Yu P.-H., Chiang C.-M., Kuo H.-K., Chen M.-H., Cheng C.-M., Van Gool L. PyNET-V2 Mobile: efficient on-device photo processing with neural networks // Proceedings of the 26th Inter. conf. on pattern recognition. 2022. pp. 677-684.</mixed-citation>
      </ref>
      <ref id="R9">
        <mixed-citation>Lee R., Venieris S.I., Dudziak L., Bhattacharya S., Lane N.D. MobiSR: Efficient On-Device Super-Resolution through Heterogeneous Mobile Processors // Mat. of the 25th Annual inter. conf. on mobile computing and networking. 2019. pp. 1-16.</mixed-citation>
      </ref>
      <ref id="R10">
        <mixed-citation>Ma N., Zhang X., Zheng H.-T., Sun J. ShuffleNet V2: Practical guidelines for efficient CNN architecture design // Computer vision ECCV 2018. 2018. pp. 116-131.</mixed-citation>
      </ref>
      <ref id="R11">
        <mixed-citation>Sandler M., Howard A., Zhu M., Zhmoginov A., Chen L.-C. MobileNetV2: Inverted residuals and linear bottlenecks // Proceedings of the IEEE/CVF conf. on computer vision and pattern recognition. 2018. pp. 4510-4520.</mixed-citation>
      </ref>
      <ref id="R12">
        <mixed-citation>Shi W., Caballero J., Huszar F., Totz J., Aitken A.P., Bishop R., Rueckert D., Wang Z. Real time single image and video super-resolution using an efficient sub-pixel convolutional neural network // Proceedings of the IEEE conf. on computer vision and pattern recognition. 2016. pp. 1874-1883.</mixed-citation>
      </ref>
      <ref id="R13">
        <mixed-citation>Zamir S.W., Arora A., Khan S., Hayat M., Khan F.S., Yang M.-H., Shao L. Learning enriched features for real image restoration and enhancement // Computer vision ECCV 2020. Lecture notes in computer science. 2020. Vol. 12370. pp. 492-511.</mixed-citation>
      </ref>
      <ref id="R14">
        <mixed-citation>Zamir S.W., Arora A., Khan S., Hayat M., Khan F.S., Yang M.-H., Shao L. Multi-stage progressive image restoration // Proceedings of the IEEE/CVF conf. on computer vision and pattern recognition. 2021. pp. 14816-14826.</mixed-citation>
      </ref>
      <ref id="R15">
        <mixed-citation>Zhang X., Zhou X., Lin M., Sun J. ShuffleNet: an extremely efficient convolutional neural network for mobile devices // Proceedings of the IEEE/CVF conf. on computer vision and pattern recognition. 2018. pp. 6848-6856.</mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>
