EmguExtensions 0.3.0
dotnet add package EmguExtensions --version 0.3.0
NuGet\Install-Package EmguExtensions -Version 0.3.0
<PackageReference Include="EmguExtensions" Version="0.3.0" />
<PackageVersion Include="EmguExtensions" Version="0.3.0" />
<PackageReference Include="EmguExtensions" />
paket add EmguExtensions --version 0.3.0
#r "nuget: EmguExtensions, 0.3.0"
#:package EmguExtensions@0.3.0
#addin nuget:?package=EmguExtensions&version=0.3.0
#tool nuget:?package=EmguExtensions&version=0.3.0
EmguExtensions
A high-performance .NET library that extends Emgu.CV (OpenCV wrapper) with zero-copy Mat
accessors, ROI utilities, async stream copy helpers, pluggable Mat compression, structured contour hierarchies, and
drawing helpers.
Features
- Zero-copy Mat access -
Span<T>,ReadOnlySpan<T>,Memory<T>,ReadOnlyMemory<T>,Memory2D<T>, andReadOnlyMemory2D<T>accessors for fast pixel manipulation and async-friendly buffer usage - ROI utilities -
Roi,SafeRoi,ConstrainRoi, andRoiFromCenterfor bounded region-of-interest cropping with configurable empty-ROI behavior - Image transforms – Rotate with adjusted bounds, letterbox creation, crop-by-bounds, and shrink-to-fit resizing that reports whether the image changed, perspective crops, reusable destinations, and coordinate mapping
- Stream copy helpers -
CopyTo(Stream)andCopyToAsync(Stream, CancellationToken)copy continuous matrices in one writing and non-continuous matrices row by row without including row padding - Mat compression - Pluggable compressors (PNG, Deflate, GZip, ZLib, Brotli, Zstd) with
CMatfor memory-efficient compressed image storage.CMat.Compressis thread-safe — concurrent calls are serialized automatically - Contour hierarchies - Structured
EmguContour,EmguContours, andEmguContourFamilywrappers for OpenCV contour trees using hierarchy links, plus area-filtered convex quadrilateral detection - Drawing helpers - Polygon geometry, multi-line text rendering with alignment, SVG path generation, and predefined colors
- Disposal infrastructure -
DisposableObject,LeaveOpenDisposableObject, andGCSafeHandlefor thread-safe resource management
Requirements
OpenCV 5 migration notes
The projects use Emgu.CV 5.0.0.6584. See the OpenCV 4 to 5 migration guide.
- Affine and rotation helpers use
Matinstead of the removed EmguMatrix<T>type. OpenCV 5's revised warping improves accuracy and can improve performance without changing these helpers' APIs; pixel-exact output can differ from OpenCV 4. DepthType.ByteCountincludes half and bfloat16, bool, uint32, uint64, and int64 storage. These sizes also support the library's raw memory and compression paths; image codecs still impose their own depth limits.Emgu.CV.FontFaceoverloads forPutTextExtended,GetTextSizeExtended, andPutTextRotatedenable Unicode and custom fonts, with wrapping and text-block alignment. Existing overloads useHersheyFonts; legacy text appearance and measurements can change.- Nearest-neighbor resize follows updated rounding rules, so old pixel baselines may need review.
Vector-backed matrices can be genuinely 1D: use
Totalfor element counts rather than assuming an Nx1 layout. - The Windows native runtime packages currently produce NU1701 metadata warnings. The test project keeps these visible without treating them as errors; Windows x64 runtime behavior is covered by the test suite.
Installation
NuGet Package Manager
Install-Package EmguExtensions
.NET CLI
dotnet add package EmguExtensions
Quick Start
Zero-Copy Mat Access
using EmguExtensions;
using Emgu.CV;
using Emgu.CV.CvEnum;
using CommunityToolkit.HighPerformance;
// Create or load a Mat
using var mat = new Mat(100, 100, DepthType.Cv8U, 1);
// Zero-copy span access for fast pixel reads/writes
var span = mat.GetSpan<byte>();
span[0] = 255; // Set first pixel
// 2D span access
var span2D = mat.GetSpan2D<byte>();
span2D[10, 20] = 128; // Row 10, Col 20
// Memory access is useful when an API needs Memory<T> or ReadOnlyMemory<T>.
ReadOnlyMemory<byte> bytes = mat.GetReadOnlyMemoryOfBytes();
Memory<byte> writableBytes = mat.GetMemoryOfBytes();
writableBytes.Span[1] = 64;
GetSpan<T>, GetMemory<T>, and their read-only counterparts require a continuous Mat. Use the 2D accessors for ROI
views or other non-continuous matrices:
using var source = new Mat(100, 100, DepthType.Cv8U, 1);
using var roi = new Mat(source, new Rectangle(10, 10, 40, 40)); // Usually non-continuous
var memory2D = roi.GetMemory2D<byte>();
memory2D.Span[0, 0] = 255;
ReadOnlyMemory2D<byte> readOnlyRoi = source.GetReadOnlyMemory2D<byte>(
new Rectangle(10, 10, 40, 40));
byte first = readOnlyRoi.Span[0, 0];
Unicode Text and Custom Fonts
Fonts are caller-owned and reusable. Font sizes are in pixels; weight: 0 uses the font's current weight.
Use an embedded font ("sans", "italic", or "uni") or pass a .ttf/.otf path to FontFace.
Available glyphs depend on the selected font.
using var font = new Emgu.CV.FontFace("uni");
using var image = EmguCvExtensions.InitMat(new Size(640, 480), channels: 3);
var white = new MCvScalar(255, 255, 255);
var size = EmguCvExtensions.GetTextSizeExtended(
"Inspeção concluída\n你好", font, 28, wrapWidth: 300);
image.PutTextExtended("Inspeção concluída\n你好", new Point(20, 60), font, 28, white,
wrapWidth: 300);
image.PutTextRotated("Etiqueta", new Point(320, 240), font, 28, white,
lineAlignment: PutTextLineAlignment.Center, angle: 30);
// Returns the largest fitting size, or zero if the minimum size cannot fit.
int chosenSize = image.PutTextInRectangle("Inspeção concluída",
new Rectangle(20, 300, 300, 100), font, white, maximumFontSize: 48);
The origin is the first line's baseline. Center and right alignment anchor the whole text block at org.X,
including wrapped lines. wrapWidth: 0 disables wrapping. PutTextInRectangle fits and clips text inside a
rectangle wholly contained in the image, preserving pixels outside it. The existing Hershey-font APIs are retained.
Perspective Cropping
Provide a strictly convex quadrilateral in top-left, top-right, bottom-right, bottom-left order. The returned image is caller-owned; the source image is preserved.
PointF[] corners = [new(40, 30), new(300, 50), new(320, 230), new(20, 210)];
using var rectified = image.CreatePerspectiveCrop(corners, new Size(280, 200));
// Without an explicit size, use the longer opposing edges, rounded up, plus one pixel.
using var inferred = image.CreatePerspectiveCrop(corners);
Corner coordinates describe pixel centers. Explicit output dimensions must each be at least two pixels.
Corners outside the source are permitted; optional borderType and borderValue control the outside samples.
Self-crossing, concave, repeated, collinear, and non-finite corners are rejected.
Ordering, Buffer Reuse, and Coordinate Mapping
PointF[] unordered = [new(320, 230), new(40, 30), new(20, 210), new(300, 50)];
var ordered = EmguCvExtensions.OrderPerspectiveCorners(unordered);
var cropSize = EmguCvExtensions.GetPerspectiveCropSize(ordered);
using var destination = new Mat();
image.PerspectiveCrop(destination, ordered, cropSize);
// Reuses destination storage when dimensions and type match; overlapping buffers are handled safely.
image.PerspectiveCrop(destination, ordered, cropSize);
using var forward = EmguCvExtensions.CreatePerspectiveTransform(ordered, cropSize);
using var inverse = EmguCvExtensions.CreatePerspectiveTransform(ordered, cropSize, inverse: true);
var cropPoints = forward.MapPerspectivePoints(ordered);
var sourcePoints = inverse.MapPerspectivePoints(cropPoints);
// Crop a contour's minimum-area rotated rectangle directly.
var rotatedRectangle = new RotatedRect(new PointF(200, 150), new SizeF(120, 80), 25);
using var rectangleCrop = image.CreatePerspectiveCrop(rotatedRectangle);
Ordering uses clockwise image coordinates, starting at the smallest X + Y; ties use the smaller Y, then X.
This convention also determines the horizontal axis for rotated rectangles, including diamond shapes. Inputs are
preserved. Transform matrices are caller-owned; mapping rejects non-finite coordinates and points at infinity.
Quadrilateral Detection
using var binary = EmguCvExtensions.InitMat(new Size(640, 480));
CvInvoke.Rectangle(binary, new Rectangle(50, 50, 300, 200), new MCvScalar(255), -1);
var candidates = binary.FindQuadrilaterals(minimumArea: 1000, approximationEpsilon: 0.02);
if (candidates.Length > 0)
{
using var largest = image.CreatePerspectiveCrop(candidates[0]);
}
Pass a nonempty single-channel Cv8U binary image; threshold or otherwise segment your image first.
Candidates are convex four-corner contour approximations, ordered and sorted by decreasing polygon area.
minimumArea and maximumArea are inclusive filters; approximationEpsilon is a fraction of contour perimeter.
External contours are used by default; retrievalMode can include nested contours. The source is preserved,
and results are managed arrays requiring no disposal. This finds geometric candidates rather than identifying documents.
Boolean Masks
using var all = EmguCvExtensions.CreateBooleanMask(image.Size, value: true);
using var selectedBytes = binary.ToByteMask();
using var selected = selectedBytes.ToBooleanMask();
using var intersection = all.MaskAnd(selected);
using var union = all.MaskOr(selected);
using var exclusive = all.MaskXor(selected);
using var inverseMask = selected.MaskNot();
using var output = EmguCvExtensions.InitMat(image.Size, channels: image.NumberOfChannels, depthType: image.Depth);
image.CopyTo(output, intersection);
// For contours or codecs that require 8-bit masks:
using var byteMask = intersection.ToByteMask();
// Row-major bool input is also supported:
using var smallMask = EmguCvExtensions.CreateBooleanMask(new bool[] { true, false, false, true }, new Size(2, 2));
Logical operations accept nonempty single-channel CvBool or Cv8U masks, including non-continuous ROI views.
Every nonzero input means true. Binary operations require matching dimensions; results are caller-owned CvBool
images containing only 0 and 1. ToByteMask produces 0 and 255. Inputs are preserved, and logical negation
keeps canonical boolean values rather than flipping all storage bits.
Safe ROI Cropping
using var source = CvInvoke.Imread("image.png");
// SafeRoi clamps to mat bounds and updates the rectangle with the effective ROI.
var roiRect = new Rectangle(-10, -10, 200, 200);
using var roi = source.SafeRoi(ref roiRect);
// Padding can be applied while still keeping the ROI inside the source bounds.
var paddedRect = new Rectangle(40, 40, 120, 80);
using var paddedRoi = source.SafeRoi(ref paddedRect, padding: 16);
// ConstrainRoi only updates the rectangle and reports whether it changed.
var candidate = new Rectangle(0, 0, 500, 500);
bool changed = source.ConstrainRoi(ref candidate);
// Empty ROI behavior is explicit when a requested ROI becomes empty.
var emptyRect = Rectangle.Empty;
using var fullSource = source.SafeRoi(ref emptyRect, EmptyRoiBehavior.CaptureSource);
// Other options:
// EmptyRoiBehavior.Default -> use OpenCV's default behavior for empty ROI
// EmptyRoiBehavior.CaptureSource -> use the full source image
// EmptyRoiBehavior.ThrowException -> throw InvalidOperationException
Image Sizing and Transforms
// Create a fixed-size letterboxed image. targetWidth and targetHeight must be positive.
using var boxed = source.CreateLetterBox(640, 480, out float scale, out int padX, out int padY);
// Shrink only when the image exceeds the target bounds. Returns true when resized.
bool wasShrunk = source.ShrinkToFitPreserveAspect(1024, 768);
// Output overloads populate dst even when no resize/rotation is needed.
using var resized = new Mat();
bool resizedWasShrunk = source.ShrinkToFitPreserveAspect(resized, 1024, 768);
using var rotated = new Mat();
source.RotateAdjustBounds(rotated, angle: 30);
Copy Mat Data to Streams
using var output = File.Create("frame.raw");
// Synchronous copy.
mat.CopyTo(output);
// Async copy with cancellation support.
await using var asyncOutput = File.Create("frame-async.raw");
await mat.CopyToAsync(asyncOutput, cancellationToken);
For continuous matrices the whole data block is written in one operation. For non-continuous matrices, such as partial-width ROI views, rows are written individually, so row padding is not included.
Mat Compression with CMat
using EmguExtensions;
// Compress a Mat for memory-efficient storage
var cmat = new CMat();
cmat.Compress(mat); // Auto-selects best storage (compressed or raw)
// Decompress back to Mat — caller owns and must dispose the returned Mat
using var restored = cmat.Decompress();
// Use a specific compressor and level
cmat.Compressor = MatCompressorBrotli.Instance;
cmat.CompressionLevel = CompressionLevel.SmallestSize;
cmat.Compress(mat);
// Compress(Mat) and Compress(MatRoi) are thread-safe — safe to call from multiple threads
Parallel.ForEach(frames, (frame, _, i) =>
{
using var mat = frame.ToMat();
compressedFrames[i].Compress(mat);
});
// Switch compressor and re-encode existing data in one step
cmat.ChangeCompressor(MatCompressorDeflate.Instance, reEncodeWithNewCompressor: true);
// Equality is hash-based (XxHash3) — O(1) for mismatches, fast for collections
bool same = cmat1 == cmat2;
// Direct compressor APIs also support async compression/decompression.
byte[] compressed = await MatCompressorBrotli.Instance.CompressAsync(mat, cancellationToken);
using var decompressed = mat.New();
await MatCompressorBrotli.Instance.DecompressAsync(compressed, decompressed, cancellationToken);
Contour Hierarchy
using EmguExtensions;
// Find contours with full hierarchy
using var contours = new EmguContours(binaryImage, RetrType.Tree);
// Iterate contour families (tree roots)
foreach (var family in contours.Families)
{
Console.WriteLine($"Area: {family.Self.Area}, Children: {family.Count}");
// Even depth = solid fill; odd depth = hole/cavity
}
Drawing Helpers
using EmguExtensions;
// Generate regular polygon vertices
var hexVertices = DrawingExtensions.GetPolygonVertices(6, new SizeF(50, 50), new PointF(100, 100));
// Multi-line text with alignment
mat.PutTextExtended("Line 1\nLine 2\nLine 3",
new Point(50, 50),
FontFace.HersheyComplex,
1.0,
EmguExtensions.WhiteColor,
lineAlignment: PutTextLineAlignment.Center);
Available Compressors
| Compressor | Class | Description |
|---|---|---|
| None | MatCompressorNone |
No compression (raw bytes) |
| PNG | MatCompressorPng |
PNG image encoding via OpenCV |
| Deflate | MatCompressorDeflate |
Deflate stream compression |
| GZip | MatCompressorGZip |
GZip stream compression |
| ZLib | MatCompressorZLib |
ZLib stream compression |
| Brotli | MatCompressorBrotli |
Brotli stream compression |
| Zstd | MatCompressorZstd |
Zstandard compression (.NET 11+) |
All compressors are singletons accessed via Instance (e.g., MatCompressorBrotli.Instance). Public CompressAsync
and DecompressAsync methods call overrideable CompressCoreAsync and DecompressCoreAsync methods, so custom
compressors can provide native async behavior instead of only using the default background-thread wrapper.
CompressionLevel values are mapped to each compressor's native level range. For Brotli, NoCompression, Fastest,
Optimal, and SmallestSize map to qualities 0, 1, 4, and 11 respectively.
CMat automatically falls back to raw (uncompressed) storage when:
- The source is smaller than
ThresholdToCompress(default 512 bytes), or - Compression produces a result larger than the original.
Benchmarks
Compressor benchmark
BenchmarkDotNet v0.15.8, Windows 11 (10.0.26200.8246/25H2/2025Update/HudsonValley2)
AMD Ryzen 9 7845HX with Radeon Graphics 3.00GHz, 1 CPU, 24 logical and 12 physical cores
.NET SDK 10.0.203
[Host] : .NET 10.0.7 (10.0.7, 10.0.726.21808), X64 RyuJIT x86-64-v4
Job-MTJJIS : .NET 10.0.7 (10.0.7, 10.0.726.21808), X64 RyuJIT x86-64-v4
MaxIterationCount=16
| Method | MatSize | CompressorName | Level | Mean | Error | StdDev | Gen0 | Gen1 | Gen2 | Allocated |
|----------- |-------- |--------------- |-------------- |------------:|------------:|------------:|---------:|---------:|---------:|----------:|
| Compress | 1920 | Brotli | NoCompression | 318.3 us | 14.55 us | 14.29 us | - | - | - | 3200 B |
| Compress | 1920 | Brotli | Fastest | 332.3 us | 54.77 us | 53.79 us | - | - | - | 2816 B |
| Compress | 1920 | None | NoCompression | 337.8 us | 11.88 us | 11.66 us | 194.3359 | 194.3359 | 194.3359 | 3686483 B |
| Compress | 1920 | Deflate | Fastest | 636.3 us | 55.72 us | 54.73 us | 1.9531 | - | - | 45744 B |
| Compress | 1920 | GZip | Fastest | 638.2 us | 28.39 us | 27.88 us | 2.4414 | - | - | 45792 B |
| Compress | 1920 | ZLib | Fastest | 640.3 us | 49.51 us | 48.62 us | 2.4414 | - | - | 45784 B |
| Compress | 1920 | GZip | NoCompression | 769.6 us | 13.23 us | 12.38 us | 276.3672 | 276.3672 | 276.3672 | 3688281 B |
| Compress | 1920 | ZLib | NoCompression | 770.1 us | 27.85 us | 27.35 us | 284.1797 | 284.1797 | 284.1797 | 3688256 B |
| Compress | 1920 | Deflate | NoCompression | 862.5 us | 14.73 us | 13.78 us | 280.2734 | 280.2734 | 280.2734 | 3688228 B |
| Compress | 1920 | Deflate | Optimal | 1,303.5 us | 42.47 us | 41.71 us | 0.9766 | - | - | 23512 B |
| Compress | 1920 | GZip | Optimal | 1,305.6 us | 93.40 us | 91.74 us | - | - | - | 23568 B |
| Compress | 1920 | ZLib | Optimal | 1,353.0 us | 67.70 us | 66.49 us | - | - | - | 23552 B |
| Compress | 1920 | Brotli | Optimal | 4,997.6 us | 181.04 us | 177.80 us | - | - | - | 1313 B |
| Compress | 1920 | GZip | SmallestSize | 8,240.1 us | 397.06 us | 389.97 us | - | - | - | 15552 B |
| Compress | 1920 | ZLib | SmallestSize | 8,380.9 us | 372.70 us | 366.04 us | - | - | - | 15536 B |
| Compress | 1920 | Deflate | SmallestSize | 8,492.4 us | 427.02 us | 419.39 us | - | - | - | 15504 B |
| Compress | 1920 | PNG | Fastest | 15,511.5 us | 1,415.48 us | 1,390.19 us | - | - | - | 25736 B |
| Compress | 1920 | PNG | NoCompression | 15,571.9 us | 654.92 us | 643.22 us | 375.0000 | 375.0000 | 375.0000 | 3694733 B |
| Compress | 1920 | PNG | Optimal | 16,101.5 us | 679.06 us | 666.93 us | - | - | - | 25496 B |
| Compress | 1920 | PNG | SmallestSize | 35,974.1 us | 2,452.16 us | 2,408.35 us | - | - | - | 9280 B |
| Compress | 1920 | Brotli | SmallestSize | 66,692.0 us | 9,677.48 us | 9,504.58 us | - | - | - | 692 B |
| | | | | | | | | | | |
| Decompress | 1920 | Deflate | SmallestSize | 113.4 us | 1.35 us | 1.19 us | - | - | - | 280 B |
| Decompress | 1920 | Deflate | Fastest | 119.7 us | 2.57 us | 2.52 us | - | - | - | 280 B |
| Decompress | 1920 | Deflate | Optimal | 123.6 us | 3.56 us | 3.50 us | - | - | - | 280 B |
| Decompress | 1920 | Deflate | NoCompression | 145.0 us | 2.05 us | 1.82 us | - | - | - | 280 B |
| Decompress | 1920 | None | NoCompression | 147.1 us | 1.93 us | 1.71 us | - | - | - | - |
| Decompress | 1920 | ZLib | Fastest | 157.1 us | 1.58 us | 1.32 us | - | - | - | 312 B |
| Decompress | 1920 | GZip | Fastest | 161.6 us | 2.30 us | 1.92 us | - | - | - | 312 B |
| Decompress | 1920 | GZip | Optimal | 162.2 us | 2.07 us | 1.93 us | - | - | - | 312 B |
| Decompress | 1920 | GZip | SmallestSize | 168.2 us | 8.01 us | 7.86 us | - | - | - | 312 B |
| Decompress | 1920 | ZLib | Optimal | 177.4 us | 1.59 us | 1.41 us | - | - | - | 312 B |
| Decompress | 1920 | GZip | NoCompression | 182.6 us | 1.61 us | 1.50 us | - | - | - | 312 B |
| Decompress | 1920 | ZLib | SmallestSize | 197.6 us | 8.23 us | 8.08 us | - | - | - | 312 B |
| Decompress | 1920 | ZLib | NoCompression | 225.5 us | 13.96 us | 13.71 us | - | - | - | 312 B |
| Decompress | 1920 | Brotli | NoCompression | 1,560.4 us | 24.39 us | 21.62 us | - | - | - | - |
| Decompress | 1920 | Brotli | Fastest | 2,107.5 us | 267.52 us | 262.74 us | - | - | - | - |
| Decompress | 1920 | Brotli | SmallestSize | 2,354.4 us | 49.89 us | 49.00 us | - | - | - | - |
| Decompress | 1920 | Brotli | Optimal | 3,384.4 us | 254.81 us | 250.26 us | - | - | - | - |
| Decompress | 1920 | PNG | Fastest | 6,079.8 us | 39.73 us | 35.22 us | - | - | - | 88 B |
| Decompress | 1920 | PNG | Optimal | 6,120.8 us | 43.09 us | 38.19 us | - | - | - | 88 B |
| Decompress | 1920 | PNG | NoCompression | 6,830.0 us | 47.14 us | 44.09 us | - | - | - | 88 B |
| Decompress | 1920 | PNG | SmallestSize | 8,797.8 us | 39.41 us | 34.94 us | - | - | - | 88 B |
Mat.ToArray() benchmark
BenchmarkDotNet v0.15.8, Windows 11 (10.0.26200.8246/25H2/2025Update/HudsonValley2)
AMD Ryzen 9 7845HX with Radeon Graphics 3.00GHz, 1 CPU, 24 logical and 12 physical cores
.NET SDK 10.0.202
[Host] : .NET 10.0.6 (10.0.6, 10.0.626.17701), X64 RyuJIT x86-64-v4
DefaultJob : .NET 10.0.6 (10.0.6, 10.0.626.17701), X64 RyuJIT x86-64-v4
| Method | MatSize | Mean | Error | StdDev | Ratio | RatioSD | Gen0 | Gen1 | Gen2 | Allocated | Alloc Ratio |
|---------------------- |-------- |-----------:|---------:|---------:|------:|--------:|---------:|---------:|---------:|----------:|------------:|
| GetRawData | 1920 | 617.7 us | 6.08 us | 5.08 us | 0.99 | 0.04 | 999.0234 | 999.0234 | 999.0234 | 3.52 MB | 1.00 |
| GetSpan.ToArray | 1920 | 623.8 us | 10.47 us | 9.28 us | 1.00 | 0.05 | 999.0234 | 999.0234 | 999.0234 | 3.52 MB | 1.00 |
| 'ToArray (extension)' | 1920 | 626.5 us | 11.84 us | 28.15 us | 1.00 | 0.06 | 999.0234 | 999.0234 | 999.0234 | 3.52 MB | 1.00 |
| | | | | | | | | | | | |
| 'ToArray (extension)' | 3840 | 1,444.4 us | 28.81 us | 32.02 us | 1.00 | 0.03 | 500.0000 | 500.0000 | 500.0000 | 14.06 MB | 1.00 |
| GetSpan.ToArray | 3840 | 1,551.3 us | 16.07 us | 15.04 us | 1.07 | 0.02 | 500.0000 | 500.0000 | 500.0000 | 14.06 MB | 1.00 |
| GetRawData | 3840 | 1,582.3 us | 21.02 us | 19.66 us | 1.10 | 0.03 | 500.0000 | 500.0000 | 500.0000 | 14.06 MB | 1.00 |
Building from Source
# Restore and build
dotnet restore
dotnet build
# Run tests
dotnet test EmguExtensions.Tests
# Pack for NuGet (Release only)
dotnet pack --configuration Release --output .
Contributing
Contributions are welcome! Please see CONTRIBUTING.md for guidelines.
License
This project is licensed under the MIT License – see the LICENSE file for details.
| Product | Versions Compatible and additional computed target framework versions. |
|---|---|
| .NET | net10.0 is compatible. net10.0-android was computed. net10.0-browser was computed. net10.0-ios was computed. net10.0-maccatalyst was computed. net10.0-macos was computed. net10.0-tvos was computed. net10.0-windows was computed. |
-
net10.0
- CommunityToolkit.HighPerformance (>= 8.4.2)
- DotNext (>= 6.9.0)
- DotNext.IO (>= 6.9.0)
- Emgu.CV (>= 5.0.0.6584)
- StageKit.Primitives (>= 0.4.1)
- System.IO.Hashing (>= 10.0.12)
NuGet packages (2)
Showing the top 2 NuGet packages that depend on EmguExtensions:
| Package | Downloads |
|---|---|
|
UVtools.Core
MSLA/DLP, file analysis, calibration, repair, conversion and manipulation |
|
|
EmguExtensions.Avalonia
Avalonia integration for EmguExtensions: convert Emgu.CV Mat images to Avalonia WriteableBitmap with direct memory locking. |
GitHub repositories
This package is not used by any popular GitHub repositories.