Matlab: Compress the file size for gif files

% ────────────────────────────────────────────────
% Compress large GIF → smaller GIF (same size, fewer colors)
% ────────────────────────────────────────────────

inputGif  = 'original.gif';      % ← change this
outputGif = 'compressed_version.gif';

[frames, map] = imread(inputGif, 'Frames', 'all');

if size(frames,3) == 3
    isRGB = true;
else
    isRGB = false;
end

nFrames = size(frames, 4);

% ── Super aggressive settings ──
nColors     = 16;          % try 12 or 8 if still too big
useDither   = false;       % no dither = smaller
delayFactor = 4.0;         % 4× slower → big size win
frameStep   = 3;           % keep only every 3rd frame (big reduction)

newFrames = cell(1, floor(nFrames / frameStep));
globalMap = [];

k = 1;
for i = 1:frameStep:nFrames
    if isRGB
        thisFrame = frames(:,:,:,i);
    else
        thisIndexed = frames(:,:,:,i);
        thisFrame   = ind2rgb(thisIndexed, map);
    end
    
    if k == 1
        if useDither
            [indexed, globalMap] = rgb2ind(thisFrame, nColors);
        else
            [indexed, globalMap] = rgb2ind(thisFrame, nColors, 'nodither');
        end
    else
        if useDither
            indexed = rgb2ind(thisFrame, globalMap);
        else
            indexed = rgb2ind(thisFrame, globalMap, 'nodither');
        end
    end
    
    newFrames{k} = indexed;
    k = k + 1;
end

% Timing – slow it down a lot
info      = imfinfo(inputGif);
origDelay = info(1).DelayTime / 100;
newDelay  = origDelay * delayFactor;
if newDelay < 0.05, newDelay = 0.05; end  % reasonable min

% Write
imwrite(newFrames{1}, globalMap, outputGif, 'gif', ...
    'LoopCount', Inf, ...
    'DelayTime', newDelay);

for kk = 2:numel(newFrames)
    imwrite(newFrames{kk}, globalMap, outputGif, 'gif', ...
        'WriteMode', 'append', ...
        'DelayTime', newDelay);
end

% Report
origSize = dir(inputGif).bytes  / 1e6;
newSize  = dir(outputGif).bytes / 1e6;
fprintf('MATLAB: %.0f MB → %.0f MB (%.0f%% smaller)\n', origSize, newSize, (origSize-newSize)/origSize*100);
disp(['Saved: ' outputGif]);

1. GIF Compressor

Matlab: Hysplit Toolkit

1. MATLAB Runs the Hysplit model
2. Mapping
3. Hysplit runs multiple datasets
4. Extract the layers from the ARL files
5. Construct Fishnet with Raster
6. RunHysplit.grids/isentropic
7. Bulit-in Function to Extract the Values from NAMS datasets

Matlab: GEOS RGB Imagery

file

Downloading the most recent imagery for 10 days.

Data: GOES-16 True Color Images over Southern Manitoba

The active operational time range for GOES-16 as GOES-East was December 18, 2017, to April 4, 2025. During this period, it provided continuous imagery and data, including ABI (Advanced Baseline Imager) bands used for RGB composites like GeoColor, which you’re interested in.code. Tool.

Matlab: GlobalHighPM₂.₅

file, 2nd. 2017/2018/2019/2020/2021/2022

Matlab: Boxplot

code

Matlab: Convert the SatPM2.5 NC to Geotiff

file

References

1. SatPM2.5 (Satellite-derived PM2.5)

MATLAB: Image Surface and Edge Boundary

File

MATLAB: Combining the wind direction/speed with air pollutant data

File.

Matlab: 本地部署Deepseek，对话模式

一、下载Ollama，修改默认安装路径

二、本地部署模型

三、MATLAB安装LLMs

四、Matlab检测代码

一、下载Ollama，修改默认安装路径

Ollama 是一个可以让你使用并管理 AI 模型的工具，相当于 R1 的承载器具，因此我们在本地部署模型前必须先安装 Ollama。打开 「Ollama 官网:」https://ollama.com/，点击“Download”。

选择适合你电脑的版本，例如“「Windows」”点击“「Download for Windows」”。

安装到指定的路径

下载完成后，双击安装，默认会装到C盘。通过如下方式修改，举例：我想安装在D:\Ollama

1. 把下载的安装包放在D:\Ollama下
2. 在D:\Ollama下，打开cmd，输入此命令敲回车即可 OllamaSetup.exe /DIR="D:\Ollama"
3. 此时就可以看到Ollama被安装到指定路径了。（如果你下载完直接点了安装，也不用担心，直接控制面板程序卸载，卸载了，重新安装即可）

二、本地部署模型

1. 部署之前，增加环境变量
增加这个环境变量的话，下载的模型会存储在此，避免占用C盘内存。 注意：增加后，重启电脑以确保生效。

2. 本地部署deepseek模型
在cmd下输入命令③代码，即可安装。



三、MATLAB安装LLMs

打开网页，添加此到当前Matlab。


四、Matlab检测代码

第一步，现在cmd里面运行部署的模型ollama run deepseek-r1:14b。

第二步，matlab中运行：file

References

• MATLAB丝滑接入deepseek！愉快摸鱼吧！[2025-03-11]
• Matlab 使用本地部署的Deepseek. [2025-03-15]

Matlab: Export the selected polygons from the Shapefile

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

% Shp Path shppath='NFDB_poly_20210707.shp'; S = readgeotable(shppath); [~, prjname, ~]=fileparts(shppath); prjpath=[prjname, '.prj']; % Example: Select polygons where 'LandType' is 'Forest' and 'Area' > 1000 selected_rows = S(S.SRC_AGENCY == "MB" & S.YEAR ==2017, :); % Loop for ii=5:10 rows=selected_rows(selected_rows.MONTH==ii, :); otpath=2017*100+ii; copyfile(prjpath, [num2str(otpath), '.prj']); otpath=[num2str(otpath), '.shp']; shapewrite(rows, otpath); end

Data: Purple Air Data

File. V2

1. https://www2.purpleair.com/
2. https://api.purpleair.com/#api-sensors-get-sensors-data
3. https://develop.purpleair.com/dashboards/organization
4. https://cyclone.unbc.ca/aqmap/data/AQCSV/

Matlab: Batch Mapping

This program requires all images and shapefiles to be in a **Geographic Coordinate System**.

Mapping with a large number of latitude and longitude points.

Matlab: MCD64A1

This program can extract burnt locations from MCD64A1 products. file.

Extract the lengths of roots from the images

File. 2nd. 3rd.

MATLAB: Layer Indices Calculation

#NDVI [2024-07-12] file.

#NDVI/ExG/GRVI/Edge/HSV/XYZ/NTSC/Lab/YCbCr [2024-07-18] file.

#NDVI/GNDVI/NDRE/ExG/GRVI/Edge/HSV/XYZ/NTSC/Lab/YCbCr [2024-08-27] file.

MATLAB: Row Detection on the Image for Segmentation Using Hough Transform

file

[1] María Pérez-Ortiz, José Manuel Peña, Pedro Antonio Gutiérrez, Jorge Torres-Sánchez, César Hervás-Martínez, Francisca López-Granados. Selecting patterns and features for between- and within- crop-row weed mapping using UAV-imagery. Expert Systems with Applications, Volume 47, 2016, 85-94.

[2] Hough Lines Transform Explained. 2017-06-05.

[3] 霍夫变换(Hough Transform)详解（附代码）. 2023-07-21.

ML: The code of K-Nearest Neighbors

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322

% Created by LI Xu % Version 1.0 % May 31, 2024 % If you have any question about this code, % please do not hesitate to contact me via E-mail: % jeremy456@163.com % Blog: % http://blog.sciencenet.cn/u/lixujeremy % http://lixuworld.blogspot.com/ clear; clc; timebegin=tic; cur_data=date; cur_time=fix(clock); str1=sprintf('%s %.2d:%.2d:%.2d', cur_data, cur_time(4), cur_time(5), cur_time(6)); fprintf('Time Begin: '); fprintf(str1); fprintf('\n'); % settings******************************************** % BackValue BackVal=0; fieldname='id'; cropname='canola'; % *************************************************** % Source Directory SouDir='input'; % Destination Directory DesDir='output'; % All images files=dir(fullfile(SouDir, "*.tif")); txtpath=fullfile(DesDir, ['note.txt']); fid=fopen(txtpath, 'w', 'n', 'US-ASCII'); % Loop for ii=1:numel(files) filename=files(ii).name; filepath=fullfile(SouDir, filename); [~, fname, ext]=fileparts(filename); cr_folder=fullfile(DesDir, fname); strname=strsplit(fname, '_'); strname=strname(end-2:end); strname=strjoin(strname, '_'); if ~isfolder(cr_folder) mkdir(cr_folder); else cmd_rmdir(cr_folder); mkdir(cr_folder); end rows=['shapefiles\BL4rows_']; plots=['shapefiles\weeds_grassy_']; wfplots=['shapefiles\wf_grassy_']; samples=['shapefiles\BL_']; rows=[rows, strname, '.shp']; plots=[plots, strname, '.shp']; wfplots=[wfplots, strname, '.shp']; samples=[samples, strname, '.tif']; disp(['[', num2str(ii), '\', num2str(numel(files)), ']~', filename]); fprintf(fid, '%s\r\n', ['[', num2str(ii), '\', num2str(numel(files)), ']~', filename]); % Clip the BL/G community from the image simagepath=fullfile(cr_folder, ['BL_', strname, '.tif']); GenClip(rows, filepath, 0, simagepath); % Convert to the ordinary image tifpath=GenOrdTifImage(simagepath); % Create the Mask maskpaths=GenCompMark(simagepath, samples); end fprintf('Time Begin: '); fprintf(str1); fprintf('\n'); cur_data=date; cur_time=fix(clock); str2=sprintf('%s %.2d:%.2d:%.2d', cur_data, cur_time(4), cur_time(5), cur_time(6)); fprintf('Time End: '); disp(str2); timespan=toc(timebegin); fprintf('Time Span: %.4f s\n', timespan); disp('*******************************************************************'); function tifpath=GenOrdTifImage(inputpath) [srcdir, fname, ~]=fileparts(inputpath); tifpath=fullfile(srcdir, [fname, '_ordinary.tif']); image=imread(inputpath); imwrite(image, tifpath); end function GenClip(oneshp, filepath, BackVal, otpath) [~, fname, ~]=fileparts(oneshp); strcmd=['gdalwarp -of GTiff -cutline ', oneshp, ' -cl ', fname, ' -crop_to_cutline ']; strcmd=[strcmd, '-dstnodata ' num2str(BackVal),' ', filepath, ' ', otpath]; [~, cmdout]=system(strcmd); end function maskpaths=GenCompMark(filepath, samples) xlspath='classes.sets.xlsx'; % Color plate uniValues=readcell(xlspath, 'Sheet', 'colorplate'); uniValues(1, :)=[]; uniValues(:, 1)=[]; [soudir, fname, ~]=fileparts(filepath); maskpaths=fullfile(soudir, [fname, '_mask.tif']); sampimage=imread(samples); try [image, geo]=readgeoraster(filepath); try info=geotiffinfo(filepath); catch info=georasterinfo(filepath); end catch image=imread(filepath); end sampimage=GenFalse(sampimage); %% Use nearest neighbor classifier mask=GenNearestNeighborClass(image, sampimage); % rendering the mask outMat=zeros(size(mask, 1), size(mask, 2), 3); outMat=RenderUniValues(mask, uniValues, outMat); showmat=[image; outMat]; % imshow(showmat); knn_euclidean=fullfile(soudir, [fname, '_knn_euclidean.png']); imwrite(showmat, knn_euclidean); try geotiffwrite(maskpaths, uint8(mask), geo); catch strcmd=['gdalinfo ', filepath]; [~, cmdout]=system(strcmd); epsg=extractBetween(cmdout, "EPSG:"," got from GeoTIFF keys"); epsg=epsg{1}; % geotiffwrite(otpath, uint8(class_imag), geo, 'GeoKeyDirectoryTag', info.GeoTIFFTags.GeoKeyDirectoryTag); geotiffwrite(maskpaths, uint8(mask), geo, 'CoordRefSysCode', ['EPSG:', epsg]); end end function output=GenFalse(input) output=[]; [rows, cols, ~]=size(input); values=unique(input(:)); values(values>=100)=[]; sample_regions=false([rows, cols, numel(values)]); % Loop to assign the matrixs for ii=1:numel(values) val=values(ii); index=find(input==val); band=sample_regions(:, :, ii); band(index)=1; sample_regions(:, :, ii)=band; end output=sample_regions; end function mask=GenNearestNeighborClass(image, sampimage) [rows, cols, ~]=size(image); % Enhance the image************************** ycbcr=rgb2ycbcr(image); ycbcr(:, :, 1)=0; ycbcr=imadjust(ycbcr, stretchlim(ycbcr), []); % imwrite(ycbcr, 'ycbcr.tif'); % ***************************************** % classes={'soil', 'canola', 'soybean'}; % nClasses=numel(classes); nClasses=size(sampimage, 3); % sample_regions=false([rows, cols, nClasses]); sample_regions=sampimage; mask=[]; % select each sample region % figure; % imshow(image); % f=figure; % for ii=1:nClasses % set(f, 'name', ['Select sample region for ', classes{ii}]); % sample_regions(:, :, ii)=roipoly(image); % end % % close(f); % Convert RGB to L*a*b colorspace lab=rgb2lab(image); % Calcualate the mean 'a*' and 'b*' value for each ROI area % a=lab(:, :, 2); % b=lab(:, :, 3); a=ycbcr(:, :, 2); b=ycbcr(:, :, 3); color_markers=repmat(0, [nClasses, 2]); for count=1:nClasses color_markers(count, 1)=mean2(a(sample_regions(:, :, count))); color_markers(count, 2)=mean2(b(sample_regions(:, :, count))); end % https://www.youtube.com/watch?v=3hEvcyCJNRc&list=PLEo-jHOqGNyUWoCSD3l3V-FjX9PnHvx5n&index=33 % Classify each pixel using the nearest neighbor rule % Each class marker now has an 'a*' and 'b*' value. % You can classify each pixel in the |lab_x| image by calculating the % Euclidean distance bewteen that pixel and each marker. The smallest % distance will tell you that the pixel most closely matched that % marker. For example, if the distance between a pixel and the read % color marker is the smallest, then the pixel would be labeled as a % red pixel. color_labels=0:nClasses-1; a=double(a); b=double(b); distance=repmat(0, [size(a), nClasses]); % Perform classification for count=1:nClasses distance(:, :, count)=((a-color_markers(count, 1)).^2+... (b-color_markers(count, 2)).^2).^0.5; end % The other formulas as follows: % https://www.saedsayad.com/k_nearest_neighbors.htm [value, label]=min(distance, [], 3); label=color_labels(label); % clear value distance colors=[0, 0, 0; 0, 255, 0; 255, 0, 0]; y=zeros(size(image)); l=double(label)+1; for m=1:rows for n=1:cols y(m, n, :)=colors(l(m, n), :); end end mask=uint8(label); end

Matlab: Convert TIF to PNG

file.

Matlab: Convert the segmented image into a labeled file

file

1. Labeling images using LabKit for semantic segmentation