python 图片识别工具源码 (微软API需密钥)

import tkinter as tk
from tkinter import filedialog, scrolledtext
from azure.ai.vision.imageanalysis import ImageAnalysisClient
from azure.ai.vision.imageanalysis.models import VisualFeatures
from azure.core.credentials import AzureKeyCredential
import threading


class ImageTranslatorApp:
    def __init__(self, master):
        self.master = master
        self.master.title("图片识别工具（需联网）")

        self.select_button = tk.Button(master, text="选择图片", command=self.select_image)
        self.select_button.grid(row=0, column=0, padx=5, pady=5)

        self.selected_image_label = tk.Label(master, text="")
        self.selected_image_label.grid(row=1, column=0, columnspan=2, padx=5, pady=5)

        self.translate_button = tk.Button(master, text="识别", command=self.translate_image)
        self.translate_button.grid(row=0, column=1, padx=5, pady=5)

        self.read_var = tk.BooleanVar(value=True)
        self.read_checkbutton = tk.Checkbutton(master, text="图片内容识别", variable=self.read_var)
        self.read_checkbutton.grid(row=2, column=1, padx=5, pady=2, sticky=tk.W)

        self.caption_var = tk.BooleanVar(value=True)
        self.caption_checkbutton = tk.Checkbutton(master, text="智能分析图片", variable=self.caption_var)
        self.caption_checkbutton.grid(row=2, column=0, padx=5, pady=2, sticky=tk.W)

        self.tags_var = tk.BooleanVar()
        self.tags_checkbutton = tk.Checkbutton(master, text="图片标签", variable=self.tags_var)
        self.tags_checkbutton.grid(row=3, column=0, padx=5, pady=2, sticky=tk.W)

        self.objects_var = tk.BooleanVar()
        self.objects_checkbutton = tk.Checkbutton(master, text="对象坐标分析", variable=self.objects_var)
        self.objects_checkbutton.grid(row=3, column=1, padx=5, pady=2, sticky=tk.W)

        self.dense_captions_var = tk.BooleanVar()
        self.dense_captions_checkbutton = tk.Checkbutton(master, text="详细对象坐标分析",
                                                         variable=self.dense_captions_var)
        self.dense_captions_checkbutton.grid(row=4, column=0, padx=5, pady=2, sticky=tk.W)

        self.smart_crops_var = tk.BooleanVar()
        self.smart_crops_checkbutton = tk.Checkbutton(master, text="智能裁剪分析", variable=self.smart_crops_var)
        self.smart_crops_checkbutton.grid(row=4, column=1, padx=5, pady=2, sticky=tk.W)

        self.people_var = tk.BooleanVar()
        self.people_checkbutton = tk.Checkbutton(master, text="人\物坐标分析", variable=self.people_var)
        self.people_checkbutton.grid(row=5, column=0, padx=5, pady=2, sticky=tk.W)

        self.textbox = scrolledtext.ScrolledText(master, height=18, width=60)
        self.textbox.grid(row=6, column=0, columnspan=2, padx=5, pady=5)
        self.client = ImageAnalysisClient(endpoint="https://这里是接口.cognitiveservices.azure.com/",
                                          credential=AzureKeyCredential("这里是密钥"))

    def select_image(self):
        file_path = filedialog.askopenfilename(filetypes=[("Image Files", "*.png;*.jpg;*.jpeg;*.gif;*.bmp")])
        if file_path:
            self.selected_image = file_path
            self.selected_image_label.config(text="已选择图片：" + self.selected_image)

    def translate_image(self):
        if hasattr(self, 'selected_image'):
            self.textbox.delete(1.0, tk.END)
            self.textbox.insert(tk.END, f"识别中，请稍等……")

            # Get visual features from the checkbuttons
            selected_features = [
                VisualFeatures.CAPTION if self.caption_var.get() else None,
                VisualFeatures.READ if self.read_var.get() else None,
                VisualFeatures.TAGS if self.tags_var.get() else None,
                VisualFeatures.OBJECTS if self.objects_var.get() else None,
                VisualFeatures.DENSE_CAPTIONS if self.dense_captions_var.get() else None,
                VisualFeatures.SMART_CROPS if self.smart_crops_var.get() else None,
                VisualFeatures.PEOPLE if self.people_var.get() else None
            ]

            # Remove None values
            selected_features = [feature for feature in selected_features if feature is not None]

            if len(selected_features) <= 0:
                self.textbox.delete(1.0, tk.END)
                self.textbox.insert(tk.END, "必须勾选一个选项进行识别。")
            else:
                threading.Thread(target=self.analyze_image, args=(selected_features,)).start()

        else:
            self.textbox.delete(1.0, tk.END)
            self.textbox.insert(tk.END, "请先选择一张图片。")

    def analyze_image(self, selected_features):
        try:
            with open(self.selected_image, 'rb') as image_file:
                image_data = image_file.read()

            result = self.client.analyze(
                image_data=image_data,
                visual_features=selected_features,
                gender_neutral_caption=False,  # Optional (default is False)
                # language="zh-Hans"
            )

            # Process result and update GUI
            self.process_result(result)

        except Exception as e:
            self.textbox.delete(1.0, tk.END)
            self.textbox.insert(tk.END, f"出现错误：{e}")

    def process_result(self, result):
        pic_orc_result = ""
        if result.caption is not None:
            pic_orc_result = f"【智能分析图片:】\n'{result.caption.text}', 准确值：{result.caption.confidence:.4f}" + "\n\n"
        if result.read is not None:
            pic_orc_result = pic_orc_result + "【图片内容识别：】\n"
            for line in result.read.blocks[0].lines:
                pic_orc_result = pic_orc_result + (f"{line.text}\n")
                # for word in line.words:
                #     print(
                #         f"     Word: '{word.text}', Bounding polygon {word.bounding_polygon}, Confidence {word.confidence:.4f}")
            pic_orc_result = pic_orc_result + "\n\n"
        if result.tags is not None:
            pic_orc_result = pic_orc_result + "【图片标签:】\n"
            for tag in result.tags.list:
                pic_orc_result = pic_orc_result + (f"{tag.name}\n")
            pic_orc_result = pic_orc_result + "\n\n"
        if result.objects is not None:
            pic_orc_result = pic_orc_result + "【对象坐标分析:】\n"
            for object in result.objects.list:
                pic_orc_result = pic_orc_result + (f"'{object.tags[0].name}', {object.bounding_box}\n")
            pic_orc_result = pic_orc_result + "\n\n"
        if result.dense_captions is not None:
            pic_orc_result = pic_orc_result + "【详细对象坐标分析:】\n"
            for caption in result.dense_captions.list:
                pic_orc_result = pic_orc_result + (f"'{caption.text}', {caption.bounding_box}\n")
            pic_orc_result = pic_orc_result + "\n\n"
        if result.smart_crops is not None:
            pic_orc_result = pic_orc_result + "【裁剪坐标及宽高比建议:】\n"
            for smart_crop in result.smart_crops.list:
                pic_orc_result = pic_orc_result + (
                    f"Aspect ratio {smart_crop.aspect_ratio}: Smart crop {smart_crop.bounding_box}\n")
            pic_orc_result = pic_orc_result + "\n\n"
        if result.people is not None:
            pic_orc_result = pic_orc_result + "【人\物坐标分析:】\n"
            for person in result.people.list:
                pic_orc_result = pic_orc_result + (f"{person.bounding_box}\n")
            pic_orc_result = pic_orc_result + "\n\n"
        if pic_orc_result == "":
            self.textbox.delete(1.0, tk.END)
            self.textbox.insert(tk.END, "未识别出任何内容，无法识别图像。")
        else:
            self.textbox.delete(1.0, tk.END)
            self.textbox.insert(tk.END, pic_orc_result)


def main():
    root = tk.Tk()
    app = ImageTranslatorApp(root)
    root.mainloop()


if __name__ == "__main__":
    main()