表情符号未用颜色渲染

我正在尝试将彩色表情符号作为叠加层添加到图像上，但表情符号似乎一直以黑白渲染，并且也是透明的：

这就是我当前正在使用的：

def add_text_overlay(image_path, text, font_path, emoji_font_path, font_size, text_spacing, x_padding, output_path,
                     text_position, offset_x, offset_y):
    with PILImage.open(image_path) as img:
        img = crop_and_resize(img, 1080, 1440)

        draw = ImageDraw.Draw(img)
        font = ImageFont.truetype(font_path, font_size)

        # Load the emoji font using PIL
        emoji_font = None
        try:
            emoji_font = ImageFont.truetype(emoji_font_path, 137)  # Adjust pixel size as needed
        except OSError as e:
            logging.error(f"Unable to load the emoji font: {e}")
            logging.error(f"Path checked: {os.path.abspath(emoji_font_path)}")
            emoji_font = font  # Fallback to regular font if emoji font fails

        lines = wrap_text(text, 28)
        fixed_line_height = font_size + text_spacing
        total_text_height = fixed_line_height * len(lines)

        initial_text_y = (img.height - total_text_height) / 2 + offset_y

        for i, line in enumerate(lines):
            text_bbox = draw.textbbox((0, 0), line, font=font)
            text_width = text_bbox[2] - text_bbox[0]

            if text_position == 'left':
                text_x = x_padding + offset_x
            elif text_position == 'right':
                text_x = img.width - text_width - x_padding + offset_x
            else:
                text_x = (img.width - text_width) / 2 + offset_x

            text_y = initial_text_y + i * fixed_line_height
            if text_x < 0:
                text_x = 0
            if text_x + text_width > img.width:
                text_x = img.width - text_width

            # Draw each character with the appropriate font
            for char in line:
                char_font = emoji_font if is_emoji(char) else font  # Use emoji font for emoji characters
                draw.text((text_x, text_y), char, font=char_font, fill="black")
                text_x += char_font.getbbox(char)[2]  # Update this line to use getbbox

        img.save(output_path)

我有“无效的像素大小”错误有一段时间，直到我将像素大小更改为 137。然后它只绘制了单色透明表情符号。

问题在于正在使用 fill="black" 为表情符号着色。为了保留表情符号的颜色，需要对文本和表情符号使用不同的绘制方法。

可以使用 PIL 的 Image.alpha_composite() 方法将表情符号作为单独的图像绘制在文本的顶部。这将保留表情符号的原始颜色和透明度。

以下是更新后的代码：

def add_text_overlay(image_path, text, font_path, emoji_font_path, font_size, text_spacing, x_padding, output_path,
                     text_position, offset_x, offset_y):
    with PILImage.open(image_path) as img:
        img = crop_and_resize(img, 1080, 1440)

        draw = ImageDraw.Draw(img)
        font = ImageFont.truetype(font_path, font_size)

        # Load the emoji font using PIL
        emoji_font = None
        try:
            emoji_font = ImageFont.truetype(emoji_font_path, 137)  # Adjust pixel size as needed
        except OSError as e:
            logging.error(f"Unable to load the emoji font: {e}")
            logging.error(f"Path checked: {os.path.abspath(emoji_font_path)}")
            emoji_font = font  # Fallback to regular font if emoji font fails

        lines = wrap_text(text, 28)
        fixed_line_height = font_size + text_spacing
        total_text_height = fixed_line_height * len(lines)

        initial_text_y = (img.height - total_text_height) / 2 + offset_y

        for i, line in enumerate(lines):
            text_x = x_padding + offset_x
            text_y = initial_text_y + i * fixed_line_height

            for char in line:
                if is_emoji(char):
                    # Draw emoji as a separate image
                    emoji_img = Image.new("RGBA", emoji_font.getsize(char))
                    emoji_draw = ImageDraw.Draw(emoji_img)
                    emoji_draw.text((0, 0), char, font=emoji_font, fill="white")  # Use white or any color you prefer for emoji

                    # Calculate emoji position and paste onto the main image
                    img.paste(emoji_img, (text_x, text_y), emoji_img)
                    text_x += emoji_img.width
                else:
                    # Draw regular text
                    draw.text((text_x, text_y), char, font=font, fill="black")
                    text_x += font.getbbox(char)[2] 

        img.save(output_path)

在这个更新后的代码中： - 我们为每个表情符号字符创建了一个新的 Image 对象。 - 我们在这个新图像上绘制了表情符号，填充颜色为白色（或喜欢的任何颜色）。 - 然后，我们使用 img.paste() 将表情符号图像粘贴到主图像上的计算位置。

此方法应该在不影响其颜色或透明度的情况下正确呈现表情符号。