diff --git a/include/boost/gil/extension/dynamic_image/image_view_factory.hpp b/include/boost/gil/extension/dynamic_image/image_view_factory.hpp
index a41dc32c5..885de7f53 100644
--- a/include/boost/gil/extension/dynamic_image/image_view_factory.hpp
+++ b/include/boost/gil/extension/dynamic_image/image_view_factory.hpp
@@ -13,8 +13,10 @@
 #include <boost/gil/dynamic_step.hpp>
 #include <boost/gil/image_view_factory.hpp>
 #include <boost/gil/point.hpp>
+#include <boost/gil/typedefs.hpp>
 #include <boost/gil/detail/mp11.hpp>
 
+#include <cstddef>
 #include <cstdint>
 
 namespace boost { namespace gil {
@@ -136,7 +138,7 @@ struct nth_channel_view_fn
 {
     using result_type = ResultView;
 
-    nth_channel_view_fn(int n) : _n(n) {}
+    nth_channel_view_fn(index_t n) : _n(n) {}
 
     template <typename View>
     auto operator()(View const& src) const -> result_type
@@ -144,7 +146,7 @@ struct nth_channel_view_fn
         return result_type(nth_channel_view(src,_n));
     }
 
-    int _n;
+    index_t _n;
 };
 
 template <typename DstP, typename ResultView, typename CC = default_color_converter>
@@ -305,7 +307,7 @@ struct nth_channel_view_type<any_image_view<Views...>>
 /// \tparam Views Models Boost.MP11-compatible list of models of ImageViewConcept
 template <typename ...Views>
 inline
-auto nth_channel_view(any_image_view<Views...> const& src, int n)
+auto nth_channel_view(any_image_view<Views...> const& src, index_t n)
     -> typename nth_channel_view_type<any_image_view<Views...>>::type
 {
     using result_view_t = typename nth_channel_view_type<any_image_view<Views...>>::type;
diff --git a/include/boost/gil/image_processing/adaptive_histogram_equalization.hpp b/include/boost/gil/image_processing/adaptive_histogram_equalization.hpp
index 1b5de4868..223e16c8b 100644
--- a/include/boost/gil/image_processing/adaptive_histogram_equalization.hpp
+++ b/include/boost/gil/image_processing/adaptive_histogram_equalization.hpp
@@ -157,9 +157,9 @@ void non_overlapping_interpolated_clahe(
     using dst_channel_t    = typename channel_type<DstView>::type;
     using coord_t          = typename SrcView::x_coord_t;
 
-    std::size_t const channels = num_channels<SrcView>::value;
-    coord_t const width        = src_view.width();
-    coord_t const height       = src_view.height();
+    index_t const channels = num_channels<SrcView>::value;
+    coord_t const width    = src_view.width();
+    coord_t const height   = src_view.height();
 
     // Find control points
 
@@ -184,7 +184,7 @@ void non_overlapping_interpolated_clahe(
 
     copy_pixels(src_view, subimage_view(view(padded_img), top_left_x, top_left_y, width, height));
 
-    for (std::size_t k = 0; k < channels; k++)
+    for (index_t k = 0; k < channels; k++)
     {
         std::vector<histogram<source_channel_t>> prev_row(new_width / tile_width_x),
             next_row((new_width / tile_width_x));
diff --git a/include/boost/gil/image_processing/convolve.hpp b/include/boost/gil/image_processing/convolve.hpp
index 91384f162..8f11bb631 100644
--- a/include/boost/gil/image_processing/convolve.hpp
+++ b/include/boost/gil/image_processing/convolve.hpp
@@ -430,7 +430,7 @@ void convolve_2d(SrcView const& src_view, Kernel const& kernel, DstView const& d
         typename color_space_type<DstView>::type
     >::value, "Source and destination views must have pixels with the same color space");
 
-    for (std::size_t i = 0; i < src_view.num_channels(); i++)
+    for (index_t i = 0; i < src_view.num_channels(); i++)
     {
         detail::convolve_2d_impl(
             nth_channel_view(src_view, i),
diff --git a/include/boost/gil/image_processing/filter.hpp b/include/boost/gil/image_processing/filter.hpp
index 1fc490232..241ea9625 100644
--- a/include/boost/gil/image_processing/filter.hpp
+++ b/include/boost/gil/image_processing/filter.hpp
@@ -124,7 +124,7 @@ void median_filter(SrcView const& src_view, DstView const& dst_view, std::size_t
             src_view.height()
         );
 
-    for (std::size_t channel = 0; channel < extended_view.num_channels(); channel++)
+    for (index_t channel = 0; channel < extended_view.num_channels(); channel++)
     {
         detail::filter_median_impl(
             nth_channel_view(extended_view, channel),
diff --git a/include/boost/gil/image_processing/histogram_equalization.hpp b/include/boost/gil/image_processing/histogram_equalization.hpp
index fba98f87c..bbf5d0aec 100644
--- a/include/boost/gil/image_processing/histogram_equalization.hpp
+++ b/include/boost/gil/image_processing/histogram_equalization.hpp
@@ -117,13 +117,13 @@ void histogram_equalization(
     using dst_channel_t    = typename channel_type<DstView>::type;
     using coord_t          = typename SrcView::x_coord_t;
 
-    std::size_t const channels = num_channels<SrcView>::value;
-    coord_t const width        = src_view.width();
-    coord_t const height       = src_view.height();
-    std::size_t pixel_max      = (std::numeric_limits<dst_channel_t>::max)();
-    std::size_t pixel_min      = (std::numeric_limits<dst_channel_t>::min)();
+    index_t const channels = num_channels<SrcView>::value;
+    coord_t const width    = src_view.width();
+    coord_t const height   = src_view.height();
+    std::size_t pixel_max  = (std::numeric_limits<dst_channel_t>::max)();
+    std::size_t pixel_min  = (std::numeric_limits<dst_channel_t>::min)();
 
-    for (std::size_t i = 0; i < channels; i++)
+    for (index_t i = 0; i < channels; i++)
     {
         histogram<source_channel_t> h;
         fill_histogram(nth_channel_view(src_view, i), h, bin_width, false, false, mask, src_mask);
diff --git a/include/boost/gil/image_processing/histogram_matching.hpp b/include/boost/gil/image_processing/histogram_matching.hpp
index 1b947c426..d254f4cd3 100644
--- a/include/boost/gil/image_processing/histogram_matching.hpp
+++ b/include/boost/gil/image_processing/histogram_matching.hpp
@@ -163,7 +163,7 @@ void histogram_matching(
     using dst_channel_t    = typename channel_type<DstView>::type;
     using coord_t          = typename SrcView::x_coord_t;
 
-    std::size_t const channels     = num_channels<SrcView>::value;
+    index_t const channels         = num_channels<SrcView>::value;
     coord_t const width            = src_view.width();
     coord_t const height           = src_view.height();
     source_channel_t src_pixel_min = (std::numeric_limits<source_channel_t>::min)();
@@ -171,7 +171,7 @@ void histogram_matching(
     ref_channel_t ref_pixel_min    = (std::numeric_limits<ref_channel_t>::min)();
     ref_channel_t ref_pixel_max    = (std::numeric_limits<ref_channel_t>::max)();
 
-    for (std::size_t i = 0; i < channels; i++)
+    for (index_t i = 0; i < channels; i++)
     {
         histogram<source_channel_t> src_histogram;
         histogram<ref_channel_t> ref_histogram;
diff --git a/include/boost/gil/image_processing/morphology.hpp b/include/boost/gil/image_processing/morphology.hpp
index babe1acb8..d6a80f497 100644
--- a/include/boost/gil/image_processing/morphology.hpp
+++ b/include/boost/gil/image_processing/morphology.hpp
@@ -112,7 +112,7 @@ void morph(SrcView const& src_view, DstView const& dst_view, Kernel const& ker_m
 
     gil::image<typename DstView::value_type> intermediate_img(src_view.dimensions());
 
-    for (std::size_t i = 0; i < src_view.num_channels(); i++)
+    for (index_t i = 0; i < src_view.num_channels(); i++)
     {
         morph_impl(nth_channel_view(src_view, i), nth_channel_view(view(intermediate_img), i),
                    ker_mat, identifier);
@@ -153,7 +153,7 @@ void difference(SrcView const& src_view1, SrcView const& src_view2, DiffView con
     gil_function_requires<ColorSpacesCompatibleConcept<
         typename color_space_type<SrcView>::type, typename color_space_type<DiffView>::type>>();
 
-    for (std::size_t i = 0; i < src_view1.num_channels(); i++)
+    for (index_t i = 0; i < src_view1.num_channels(); i++)
     {
         difference_impl(nth_channel_view(src_view1, i), nth_channel_view(src_view2, i),
                         nth_channel_view(diff_view, i));
diff --git a/include/boost/gil/image_processing/threshold.hpp b/include/boost/gil/image_processing/threshold.hpp
index b3b56356f..7b106abc0 100644
--- a/include/boost/gil/image_processing/threshold.hpp
+++ b/include/boost/gil/image_processing/threshold.hpp
@@ -337,7 +337,7 @@ void threshold_optimal
 {
     if (mode == threshold_optimal_value::otsu)
     {
-        for (std::size_t i = 0; i < src_view.num_channels(); i++)
+        for (index_t i = 0; i < src_view.num_channels(); i++)
         {
             detail::otsu_impl
                 (nth_channel_view(src_view, i), nth_channel_view(dst_view, i), direction);
diff --git a/include/boost/gil/image_view_factory.hpp b/include/boost/gil/image_view_factory.hpp
index 73c2da6ec..57ba59260 100644
--- a/include/boost/gil/image_view_factory.hpp
+++ b/include/boost/gil/image_view_factory.hpp
@@ -14,6 +14,7 @@
 #include <boost/gil/image_view.hpp>
 #include <boost/gil/metafunctions.hpp>
 #include <boost/gil/point.hpp>
+#include <boost/gil/typedefs.hpp>
 #include <boost/gil/detail/mp11.hpp>
 
 #include <boost/assert.hpp>
@@ -338,7 +339,7 @@ namespace detail {
     struct __nth_channel_view_basic<View,false> {
         using type = typename view_type<typename channel_type<View>::type, gray_layout_t, false, true, view_is_mutable<View>::value>::type;
 
-        static type make(View const& src, int n) {
+        static type make(View const& src, index_t n) {
             using locator_t = typename type::xy_locator;
             using x_iterator_t = typename type::x_iterator;
             using x_iterator_base_t = typename iterator_adaptor_get_base<x_iterator_t>::type;
@@ -351,7 +352,7 @@ namespace detail {
     template <typename View>
     struct __nth_channel_view_basic<View,true> {
         using type = typename view_type<typename channel_type<View>::type, gray_layout_t, false, false, view_is_mutable<View>::value>::type;
-        static type make(View const& src, int n) {
+        static type make(View const& src, index_t n) {
             using x_iterator_t = typename type::x_iterator;
             return interleaved_view(src.width(),src.height(),(x_iterator_t)&(src(0,0)[n]), src.pixels().row_size());
         }
@@ -375,7 +376,7 @@ namespace detail {
     public:
         using type = typename __nth_channel_view_basic<View,adjacent>::type;
 
-        static type make(View const& src, int n) {
+        static type make(View const& src, index_t n) {
             return __nth_channel_view_basic<View,adjacent>::make(src,n);
         }
     };
@@ -403,11 +404,11 @@ namespace detail {
         using reference = mp11::mp_if_c<is_mutable, ref_t, value_type>;
         using result_type = reference;
 
-        nth_channel_deref_fn(int n=0) : _n(n) {}
+        nth_channel_deref_fn(index_t n=0) : _n(n) {}
         template <typename P>
         nth_channel_deref_fn(const nth_channel_deref_fn<P>& d) : _n(d._n) {}
 
-        int _n;        // the channel to use
+        index_t _n;        // the channel to use
 
         auto operator()(argument_type srcP) const -> result_type
         {
@@ -421,7 +422,7 @@ namespace detail {
         using AD = typename View::template add_deref<deref_t>;
     public:
         using type = typename AD::type;
-        static type make(View const& src, int n) {
+        static type make(View const& src, index_t n) {
             return AD::make(src, deref_t(n));
         }
     };
@@ -440,12 +441,12 @@ struct nth_channel_view_type {
     using VB = detail::__nth_channel_view<View,view_is_basic<View>::value>;
 public:
     using type = typename VB::type;
-    static type make(View const& src, int n) { return VB::make(src,n); }
+    static type make(View const& src, index_t n) { return VB::make(src,n); }
 };
 
 /// \ingroup ImageViewTransformationsNthChannel
 template <typename View>
-typename nth_channel_view_type<View>::type nth_channel_view(View const& src, int n) {
+typename nth_channel_view_type<View>::type nth_channel_view(View const& src, index_t n) {
     return nth_channel_view_type<View>::make(src,n);
 }
 
@@ -454,11 +455,11 @@ typename nth_channel_view_type<View>::type nth_channel_view(View const& src, int
 /// \brief single-channel (grayscale) view of the K-th channel of a given image_view. The channel index is a template parameter
 
 namespace detail {
-    template <int K, typename View, bool AreChannelsTogether> struct __kth_channel_view_basic;
+    template <index_t K, typename View, bool AreChannelsTogether> struct __kth_channel_view_basic;
 
     // kth_channel_view when the channels are not adjacent in memory. This can happen for multi-channel interleaved images
     // or images with a step
-    template <int K, typename View>
+    template <index_t K, typename View>
     struct __kth_channel_view_basic<K,View,false> {
     private:
         using channel_t = typename kth_element_type<typename View::value_type,K>::type;
@@ -475,7 +476,7 @@ namespace detail {
     };
 
     // kth_channel_view when the channels are together in memory (true for simple grayscale or planar images)
-    template <int K, typename View>
+    template <index_t K, typename View>
     struct __kth_channel_view_basic<K,View,true> {
     private:
         using channel_t = typename kth_element_type<typename View::value_type, K>::type;
@@ -487,10 +488,10 @@ namespace detail {
         }
     };
 
-    template <int K, typename View, bool IsBasic> struct __kth_channel_view;
+    template <index_t K, typename View, bool IsBasic> struct __kth_channel_view;
 
     // For basic (memory-based) views dispatch to __kth_channel_view_basic
-    template <int K, typename View> struct __kth_channel_view<K,View,true>
+    template <index_t K, typename View> struct __kth_channel_view<K,View,true>
     {
     private:
         using src_x_iterator = typename View::x_iterator;
@@ -515,7 +516,7 @@ namespace detail {
     /// If the input is a pixel value or constant reference, the function object is immutable. Otherwise it is mutable (and returns non-const reference to the k-th channel)
     /// \tparam SrcP reference to PixelConcept (could be pixel value or const/non-const reference)
     /// Examples: pixel<T,L>, pixel<T,L>&, const pixel<T,L>&, planar_pixel_reference<T&,L>, planar_pixel_reference<const T&,L>
-    template <int K, typename SrcP>
+    template <index_t K, typename SrcP>
     struct kth_channel_deref_fn
     {
         static constexpr bool is_mutable =
@@ -544,7 +545,7 @@ namespace detail {
         }
     };
 
-    template <int K, typename View> struct __kth_channel_view<K,View,false> {
+    template <index_t K, typename View> struct __kth_channel_view<K,View,false> {
     private:
         using deref_t = kth_channel_deref_fn<K,typename View::reference>;
         using AD = typename View::template add_deref<deref_t>;
@@ -562,7 +563,7 @@ namespace detail {
 /// If the channels in the source view are adjacent in memory (such as planar non-step view or single-channel view) then the
 /// return view is a single-channel non-step view.
 /// If the channels are non-adjacent (interleaved and/or step view) then the return view is a single-channel step view.
-template <int K, typename View>
+template <index_t K, typename View>
 struct kth_channel_view_type {
 private:
     BOOST_GIL_CLASS_REQUIRE(View, boost::gil, ImageViewConcept)
@@ -573,7 +574,7 @@ struct kth_channel_view_type {
 };
 
 /// \ingroup ImageViewTransformationsKthChannel
-template <int K, typename View>
+template <index_t K, typename View>
 auto kth_channel_view(View const& src)
     -> typename kth_channel_view_type<K,View>::type
 {
diff --git a/include/boost/gil/typedefs.hpp b/include/boost/gil/typedefs.hpp
index 44673954c..82470d5e8 100644
--- a/include/boost/gil/typedefs.hpp
+++ b/include/boost/gil/typedefs.hpp
@@ -1,6 +1,7 @@
 //
 // Copyright 2005-2007 Adobe Systems Incorporated
 // Copyright 2018 Mateusz Loskot <mateusz@loskot.net>
+// Copyright 2022 Dirk Stolle
 //
 // Use, modification and distribution are subject to the Boost Software License,
 // Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
@@ -16,6 +17,7 @@
 #include <boost/gil/rgb.hpp>
 #include <boost/gil/rgba.hpp>
 
+#include <cstddef>
 #include <cstdint>
 #include <memory>
 #if !defined(BOOST_NO_CXX17_HDR_MEMORY_RESOURCE)
@@ -120,6 +122,9 @@ template <typename B, typename Mn, typename Mx> struct scoped_channel_value;
 template <typename T> struct float_point_zero;
 template <typename T> struct float_point_one;
 
+// general-purpose index type
+using index_t = std::ptrdiff_t;
+
 //////////////////////////////////////////////////////////////////////////////////////////
 ///  Built-in channel models
 //////////////////////////////////////////////////////////////////////////////////////////