Creating a Cherry Blossom Timelapse with Python: Image Processing and Video Generation

The post introduces a special project launched by Wuhan University to livestream cherry blossom scenes and explains how a student used Python to simulate the blossom opening process, which later became a popular online showcase.

First, each video frame is saved as a JPEG image using OpenCV's cv2.imwrite("pic/frame%d.jpg" % count, image) # save frame as JPEG file .

Next, a blank canvas is created with Pillow, a font is loaded, and a nested loop iterates over the image pixels. For each sampled pixel, the corresponding color is used to set the drawing ink and render the Chinese phrase "武汉加油" on the canvas:

<code>blank = Image.new("RGB", [len(img[0]), len(img)], "white")
drawObj = ImageDraw.Draw(blank)

n = 10
font = ImageFont.truetype("C:/Windows/Fonts/Microsoft YaHei UI/msyhbd.ttc", size=n-1)

for i in range(0, len(img), n):
    for j in range(0, len(img[i]), n):
        text = "武汉加油"
        drawObj.ink = img[i][j][0] + img[i][j][1]*256 + img[i][j][2]*256*256
        drawObj.text([j, i], text[int(j/n) % len(text)], font=font)
        print("完成处理——", i, j)

blank.save("new/new_" + pic, "jpeg")
</code>

Finally, the generated images are compiled into a video file. The picvideo function reads all JPEG frames, sets the video codec, frame rate, and resolution, and writes each frame to an MP4 file:

<code>def picvideo(path, size):
    filelist = os.listdir(path)
    filelist = resort(filelist)
    fps = 24
    file_path = "video/new.mp4"
    fourcc = cv2.VideoWriter_fourcc('D','I','V','X')
    video = cv2.VideoWriter(file_path, fourcc, fps, size)
    for item in filelist:
        if item.endswith('.jpg'):
            img = cv2.imread(os.path.join(path, item))
            video.write(img)
    video.release()
</code>

The resulting video shows thousands of "武汉加油" characters forming a blooming cherry blossom, symbolizing support for Wuhan and demonstrating practical image‑processing techniques with Python.